Not in the top 1,000 by GitHub stars
NoobyGains/godmode
Engineering discipline for AI coding agents: question the request, verify against current docs, plan, build test-first on the cheapest capable model, review with a second model family, prove it before saying done. One repo for Claude Code, Codex, Antigravity, Cursor, Gemini CLI and OpenCode.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
skills/agent-briefing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/code-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/codebase-conventions/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/completion-gate/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/cross-model-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/delegated-execution/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/doctor/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/environment-awareness/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/error-recovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/fault-diagnosis/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/grounding/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/inline-execution/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/intent-discovery/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/knowledge-capture/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/merge-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/model-routing/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/parallel-execution/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/performance-tuning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/project-bootstrap/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/quality-gates/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/rationale/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/reference-first/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/review-response/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/security-protocol/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/system-design/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/task-planning/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/team-orchestration/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/test-first/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/ui-engineering/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/ui-research/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/using-godmode/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/workspace-isolation/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
skills/writing-skills/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
ef8dbf444dc7 → ef8dbf444dc7
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → ef8dbf444dc7
No earlier version was observed for comparison.
Verified commit ef8dbf444dc7c854a77139d333471ee343181458
git clone --filter=blob:none --no-checkout https://github.com/NoobyGains/godmode.git .hotskills/noobygains-godmode && git -C .hotskills/noobygains-godmode checkout --detach ef8dbf444dc7c854a77139d333471ee343181458
# Copy each selected skill directory from .hotskills/noobygains-godmode into .claude/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/NoobyGains/godmode.git .hotskills/noobygains-godmode && git -C .hotskills/noobygains-godmode checkout --detach ef8dbf444dc7c854a77139d333471ee343181458
# Copy each selected skill directory from .hotskills/noobygains-godmode into .agents/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/NoobyGains/godmode.git .hotskills/noobygains-godmode && git -C .hotskills/noobygains-godmode checkout --detach ef8dbf444dc7c854a77139d333471ee343181458
# Copy each selected skill directory from .hotskills/noobygains-godmode into .opencode/skills/<skill-name>/Heat
5.8- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
- agent-briefing
skills/agent-briefing - code-review
skills/code-review - codebase-conventions
skills/codebase-conventions - completion-gate
skills/completion-gate - cross-model-review
skills/cross-model-review - delegated-execution
skills/delegated-execution - doctor
skills/doctor - environment-awareness
skills/environment-awareness - error-recovery
skills/error-recovery - fault-diagnosis
skills/fault-diagnosis - grounding
skills/grounding - inline-execution
skills/inline-execution - intent-discovery
skills/intent-discovery - knowledge-capture
skills/knowledge-capture - merge-protocol
skills/merge-protocol - model-routing
skills/model-routing - parallel-execution
skills/parallel-execution - performance-tuning
skills/performance-tuning - project-bootstrap
skills/project-bootstrap - quality-gates
skills/quality-gates - rationale
skills/rationale - reference-first
skills/reference-first - review-response
skills/review-response - security-protocol
skills/security-protocol - system-design
skills/system-design - task-planning
skills/task-planning - team-orchestration
skills/team-orchestration - test-first
skills/test-first - ui-engineering
skills/ui-engineering - ui-research
skills/ui-research - using-godmode
skills/using-godmode - workspace-isolation
skills/workspace-isolation - writing-skills
skills/writing-skills
How it works
agent-briefing
When your agent uses it
Use when writing a prompt for a subagent or teammate, deciding how much context another agent needs, writing or reading a handoff report between agents, or sending a message to another agent mid-task.
What it loads into context
Frontmatter
50 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Brief
- Lean Versus Full Context
- The Report-File Rule
- Handoff Report Formats
- Messages Between Agents
- The Protocol
- Cognitive Traps
- Guardrails
- Connections
Files
../../.claude-plugin/marketplace.jsonasset900 bytes../../.claude-plugin/plugin.jsonasset921 bytes../../assets/icon.pngasset264,630 bytes../../hooks/hooks.jsonasset560 bytesSKILL.mdcore9,088 bytes
Read SKILL.md
# Agent Briefing
Most multi-agent failures are briefing failures: the worker lacked one fact, drowned in irrelevant history, guessed at a constraint, or returned a report nobody could check. A dispatched agent knows only what its brief says, and you know only what its report says. Both handoffs lose detail unless they are written on purpose.
**Core principle: write every brief for a capable stranger with no memory of this session, and require every report as a file whose claims can be checked.** Give exactly the context the task needs, point to files instead of pasting them, and ask for evidence, not reassurance.
## The Prime Directive
```
NO DISPATCH WITHOUT A COMPLETE BRIEF, NO HANDOFF WITHOUT A REPORT FILE
```
No exceptions. No workarounds. No shortcuts.
View the complete source on GitHubcode-review
When your agent uses it
Use when a task, feature or bug fix is implemented and before it is committed, merged or handed back to the user, and when the user asks for a review of a change, a branch or a pull request.
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
2,013 tokensWhen the skill is activated
Reference documents
988 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- What the Reviewer Gets, and What It Does Not
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore9,069 bytesSKILL.mdcore9,069 bytesreviewer-template.mddoc4,277 bytes
Read SKILL.md
# Code Review
A change is reviewed by someone who did not write it, in two stages: first "does it do exactly what was asked", then "is it built well". The author then explains the change to the user in plain language and discloses everything that was not asked for. **The author's confidence is not evidence; a fresh reviewer reading the diff is.**
## The Prime Directive
```
NO CHANGE LANDS UNTIL A FRESH REVIEWER HAS PASSED IT AND THE USER HAS SEEN WHAT CHANGED
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubcodebase-conventions
When your agent uses it
Use when about to add or change code in an existing codebase: a new file, module, route, component, service, test, migration, error path or dependency, especially in a project the agent did not write or has not yet surveyed in this session.
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
2,612 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- The Convention Catalogue
- Typical Drift
- When No Precedent Exists
- When the Convention Looks Wrong
- Pre-Commit Conformity Check
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore11,531 bytes
Read SKILL.md
# Codebase Conventions
An existing codebase is the most authoritative reference for how new code in it should look. Before adding anything, find the two or three closest precedents, write down the conventions they share, build the new code by shadowing the nearest one, and check the result against them before committing. **New code must read as if the original authors wrote it: one way of doing each thing, not two.**
## The Prime Directive
```
NO NEW CODE WITHOUT A NAMED PRECEDENT IN THIS REPOSITORY
```
No exceptions. No workarounds. No shortcuts.
If you cannot point to the file you modelled your change on, you guessed. When no precedent exists, say so, and establish the convention deliberately.
View the complete source on GitHubcompletion-gate
When your agent uses it
Use when about to say that work is done, fixed, passing, working, deployed or ready, before committing, opening or merging a pull request, closing a task, moving to the next task, or reporting a subagent's result as finished.
What it loads into context
Frontmatter
307 tokensAlways available to the agent
SKILL.md
2,370 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- What Counts as Proof
- Configuration Changes
- UI Changes
- Reporting Honestly
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,271 bytes
Read SKILL.md
# Completion Gate
A claim about the state of the work is a statement of fact, and a fact needs evidence. The evidence is output you produced and read in this message: a test run, a build, a request and its response, a screenshot. **If you did not run it just now, you do not know it.**
## The Prime Directive
```
NO CLAIM OF DONE, FIXED OR PASSING WITHOUT FRESH VERIFICATION OUTPUT IN THIS MESSAGE
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubcross-model-review
When your agent uses it
Use when a reviewed change is about to land on a shared branch, be released or be deployed, and when the user asks for a second opinion on a change from a different model or a different AI vendor.
What it loads into context
Frontmatter
51 tokensAlways available to the agent
SKILL.md
2,084 tokensWhen the skill is activated
Reference documents
648 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Writing Good Questions
- Keeping It Cheap
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore9,297 bytesreview-brief-template.mddoc2,589 bytes
Read SKILL.md
# Cross-Model Review
Models from one family share blind spots: the same training habits, the same favourite mistakes, the same things they never question. Before a change lands, one reviewer from a **different model family** than the author reads it once, narrowly, from a brief file. **One focused outside read catches what a whole team of same-family reviewers will agree to miss.**
## The Prime Directive
```
NOTHING LANDS WITHOUT ONE READ-ONLY REVIEW FROM A DIFFERENT MODEL FAMILY, OR A RECORDED REASON WHY NOT
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubdelegated-execution
When your agent uses it
Use when a written implementation plan is ready to execute in the current session and its tasks are mostly independent, so each one can go to a fresh implementer subagent with its own review.
What it loads into context
Frontmatter
47 tokensAlways available to the agent
SKILL.md
3,167 tokensWhen the skill is activated
Reference documents
4,940 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Operating Rules
- The Protocol
- The Task Loop
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore13,644 bytesimplementer-prompt.mddoc4,491 bytesre-review-prompt.mddoc2,479 bytesreferences/ledger-and-workspace.mddoc7,645 bytesreviewer-prompt.mddoc4,954 bytes
Read SKILL.md
# Delegated Execution
Execute a plan by giving each task to a fresh implementer subagent, reviewing each task with a fresh reviewer, and closing with one review of the whole branch. You are the controller: you prepare briefs, dispatch, read short reports, rule on conflicts, and keep the ledger. You do not write the code.
**Core principle: the ledger on disk is the memory, the brief is the requirement, the report file is the evidence, and a fresh reviewer is the gate.** Your own recollection is none of these; after a context compaction it is the least reliable thing in the session.
## The Prime Directive
```
NO TASK IS COMPLETE UNTIL A FRESH REVIEWER PASSES IT AND THE LEDGER SAYS SO
```
No exceptions. No workarounds. No shortcuts.
View the complete source on GitHubdoctor
When your agent uses it
Use when GodMode skills do not seem to trigger, when the user asks whether GodMode is installed or working, or after installing or updating the plugin on any harness
What it loads into context
Frontmatter
48 tokensAlways available to the agent
SKILL.md
826 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- What the script checks
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore3,619 bytes
Read SKILL.md
# Doctor
Checks the GodMode installation from inside the running harness and reports what it finds. **The bootstrap that makes skills trigger is injected at session start; when it is missing, every other skill is inert, so this check comes first.**
## The Prime Directive
```
DIAGNOSE THE INSTALL BEFORE BLAMING THE SKILLS
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubenvironment-awareness
When your agent uses it
Use when about to run the first shell command in a session or a new project, when installing dependencies, when a command fails with not found or not recognised, when writing platform-specific code or paths, or when entering a container, virtual machine, remote host or CI runner.
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
2,842 tokensWhen the skill is activated
Reference documents
1,960 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Lockfiles and Managers
- Windows Specifics
- When to Probe and When to Assume
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore11,756 bytesreferences/platform-commands.mddoc6,410 bytes
Read SKILL.md
# Environment Awareness
Most wasted commands in an agent session come from a wrong guess about where the command runs: Unix syntax in PowerShell, npm in a pnpm project, a global Python instead of the project's virtual environment. You find out where you are with a few cheap probes and then write every command for that place. **Probe once, then act on what you saw, not on what is typical.**
## The Prime Directive
```
NO SHELL COMMAND UNTIL YOU KNOW THE OS, THE SHELL AND THE PACKAGE MANAGER
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHuberror-recovery
When your agent uses it
Use when a fix attempt has failed more than once for the same problem, when a fix for one problem causes another, when the same file keeps being edited without progress, when test results get worse, or when the work feels stuck in a loop.
What it loads into context
Frontmatter
59 tokensAlways available to the agent
SKILL.md
2,205 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- The Severity Ladder
- Fundamentally Different, Not a Variation
- Rolling Back
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore9,609 bytesSKILL.mdcore9,609 bytes
Read SKILL.md
# Error Recovery
Agents get stuck in a recognisable way: the same approach, retried with small variations, each attempt adding code and confidence but not progress. This skill counts the attempts, escalates at fixed thresholds and forces either a genuinely different approach, a rollback, or an honest report to the user. **Retries are bounded; the counter decides when to stop, not your optimism.**
## The Prime Directive
```
EVERY ATTEMPT IS COUNTED, AND AFTER THE THIRD FAILURE YOU STOP AND TELL THE USER
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubfault-diagnosis
When your agent uses it
Use when a test fails, a build breaks, a bug is reported, behaviour is unexpected, performance degrades or an integration stops working, before proposing or making any fix.
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
2,187 tokensWhen the skill is activated
Reference documents
2,771 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- When There Is No Root Cause in the Code
- Signals That You Have Skipped a Phase
- Cognitive Traps
- Guardrails
- Connections
Scripts executed separately
references/condition-based-waiting-example.ts
Files
SKILL.mdcore9,650 bytesSKILL.mdcore9,650 bytesreferences/condition-based-waiting-example.tsruns3,205 bytesreferences/condition-based-waiting.mddoc3,170 bytesreferences/defense-in-depth.mddoc3,637 bytesreferences/root-cause-tracing.mddoc5,059 bytes
Read SKILL.md
# Fault Diagnosis
A fix that is not aimed at the root cause is a guess, and guesses pile up: each one adds a change, hides a symptom and makes the next failure harder to read. Diagnosis runs in four phases (reproduce, isolate, find the root cause, fix with a regression test) and each phase leaves evidence behind. **No fix until you can say what is broken, where, and why.**
## The Prime Directive
```
NO FIX WITHOUT A REPRODUCED FAILURE AND A ROOT CAUSE YOU CAN STATE
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubgrounding
When your agent uses it
Use when about to write code, config or commands against a library, API, CLI, SDK, model or platform, or to state a version, flag, limit, price, licence, release date or best practice to the user, or when an error suggests an interface has changed
What it loads into context
Frontmatter
64 tokensAlways available to the agent
SKILL.md
1,946 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Facts That Can Change
- What Counts as a Primary Source
- The Protocol
- Recording Sources
- When Sources Conflict or Are Missing
- Proportionality
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytesSKILL.mdcore8,492 bytes
Read SKILL.md
# Grounding
What you remember about an API, a CLI flag, a version or a price was true at some point before your training ended, and may not be true now. **Treat every fact that can change as a hypothesis: verify it against a current primary source, record where and when you checked, and let the source win when it disagrees with memory.**
## The Prime Directive
```
NO FACT THAT CAN CHANGE GOES IN CODE OR AN ANSWER UNVERIFIED
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubinline-execution
When your agent uses it
Use when a written implementation plan is to be executed by the current agent itself rather than by subagents: the user chose the cheaper mode, the harness cannot dispatch subagents, or the plan is short and fully specified.
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
2,565 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Operating Rules
- The Protocol
- The Task Loop
- Final Review
- Finish
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,999 bytes
Read SKILL.md
# Inline Execution
Execute the plan yourself, task by task, in this context. There is no implementer or reviewer per task; there is one fresh review of the whole branch at the end. This is the cheapest execution mode: one context plus one reviewer.
**Core principle: the plan already did the thinking, so execute it exactly, prove every step with a test you watched fail and pass, and leave a ledger that survives your own forgetting.** Inline mode gives up a fresh context and a second pair of eyes per task. It buys them back by other means: the brief is the spec, the ledger is the memory, test-first is the per-task gate, and the final reviewer is the second pair of eyes.
## The Prime Directive
```
NO TASK IS DONE WITHOUT A TEST RUN YOU READ AND A LEDGER LINE THAT RECORDS IT
```
No exceptions. No workarounds. No shortcuts.
View the complete source on GitHubintent-discovery
When your agent uses it
Use when the user wants something new built or existing behaviour changed, such as a feature, component, page, API, module or project, and no approved design for it exists yet in this conversation or in docs/plans
What it loads into context
Frontmatter
54 tokensAlways available to the agent
SKILL.md
2,497 tokensWhen the skill is activated
Reference documents
481 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- The Design Document
- The Specification Section
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,815 bytesreferences/builder-check-prompt.mddoc2,158 bytes
Read SKILL.md
# Intent Discovery
Turn a request into a design the user has approved, with a specification precise enough that a capable builder could plan from it without asking again. You build what you believe the user wants, and that belief is usually thinner than it feels. **Find out what they want and why, one question at a time, and get sign-off on a written design before anything is built.**
## The Prime Directive
```
NOTHING GETS BUILT UNTIL THE USER APPROVES THE DESIGN
```
No exceptions. No workarounds. No shortcuts.
Building includes writing product code, scaffolding, installing dependencies, creating projects and invoking any implementation skill. Reading and exploring are fine.
View the complete source on GitHubknowledge-capture
When your agent uses it
Use when a substantial task finishes, when the user corrects the agent or states a preference, when a non-obvious cause or project convention is discovered the hard way, or when the user asks the agent to remember or forget something
What it loads into context
Frontmatter
56 tokensAlways available to the agent
SKILL.md
1,696 tokensWhen the skill is activated
Reference documents
2,667 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Where Memory Lives
- The Protocol
- Entry Format
- What Good Lessons Look Like
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore7,293 bytes../using-godmode/references/claude-code-tools.mddoc10,182 bytes
Read SKILL.md
# Knowledge Capture
Each session starts cold. What you learned the hard way in this one (a hidden convention, a root-cause signature, a preference the user stated) is lost unless it is written where the next session will read it. **Save durable, non-obvious lessons into the harness's own memory or the project's docs, never into a store of your own invention, and keep that memory short, current and free of duplicates.**
## The Prime Directive
```
SAVE WHAT THE NEXT SESSION CANNOT REDISCOVER CHEAPLY, WHERE IT WILL BE READ
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubmerge-protocol
When your agent uses it
Use when implementation on a branch is finished and reviewed and the work needs to land, be proposed as a pull request, be parked for later, or be thrown away.
What it loads into context
Frontmatter
43 tokensAlways available to the agent
SKILL.md
1,948 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore7,849 bytesSKILL.mdcore7,849 bytesSKILL.mdcore7,849 bytesSKILL.mdcore7,849 bytes
Read SKILL.md
# Merge Protocol
The last step of a run decides what happens to the branch: merge it locally, open a pull request, keep it as it is, or abandon it. Each path is easy to get slightly wrong (merging into the wrong base, deleting a worktree that held the only copy of a file, force-pushing over someone's commits), and most of those mistakes are expensive to undo.
**Core principle: prove the suite is green on the exact tree you are about to land, let the user choose the path, execute it exactly, and tear down only what this run created.** Integration is the user's decision; your job is to make each option safe.
## The Prime Directive
```
NO LANDING ON A RED SUITE, NO ABANDONING WITHOUT THE TYPED WORD
```
No exceptions. No workarounds. No shortcuts.
View the complete source on GitHubmodel-routing
When your agent uses it
Use when dispatching any subagent, teammate or reviewer, choosing a model or effort level for a piece of delegated work, escalating after an agent fails, or when a run's token or quota budget matters.
What it loads into context
Frontmatter
50 tokensAlways available to the agent
SKILL.md
2,044 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Tiers
- Seat Defaults
- The Protocol
- Signals That a Tier Is Wrong
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore9,053 bytes
Read SKILL.md
# Model Routing
Every seat in a run (implementer, reviewer, investigator, lead) needs a model and an effort level. Left unnamed, a dispatch inherits the session's model, which is usually the most capable and most expensive one, and the cost discipline of the whole run silently disappears.
[Credential-related source omitted]
<!-- Model order facts: https://code.claude.com/docs/en/sub-agents and https://code.claude.com/docs/en/agent-teams (fetched 2026-10-25): an omitted per-dispatch model falls back to the agent definition, then a configured default, then the main session's model; teammates fall back to the lead's model, and teammates use significantly more tokens than subagents. Concrete model names, effort flags and how to pass them per harness: skills/using-godmode/references/<harness>-tools.md. -->
## The Prime Directive
```
NO DISPATCH WITHOUT A NAMED MODEL AND EFFORT
```
View the complete source on GitHubparallel-execution
When your agent uses it
Use when two or more problems can each be solved without the others' results or shared state, such as unrelated test failures in separate subsystems, independent investigations, or separate modules with no common files.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
1,799 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Brief Skeleton
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore8,001 bytes
Read SKILL.md
# Parallel Execution
When several problems are genuinely independent, working them one after another wastes wall-clock time and fills one context with unrelated detail. Give each problem its own agent with a focused brief, let them run at the same time, and reconcile the results yourself.
**Core principle: one agent per independent problem domain, independence proven before dispatch, and a full test run after reconciliation.** Parallelism is only safe when nothing is shared; the check comes first, not after the merge conflict.
## The Prime Directive
```
NO PARALLEL DISPATCH UNTIL INDEPENDENCE IS PROVEN AND ISOLATION IS IN PLACE
```
No exceptions. No workarounds. No shortcuts.
View the complete source on GitHubperformance-tuning
When your agent uses it
Use when something is reported slow, sluggish, memory-hungry or too large, when a performance budget, Core Web Vitals threshold or latency target is breached, when asked to optimise or make code faster, or when tempted to add a cache, index, async rewrite or micro-optimisation.
What it loads into context
Frontmatter
69 tokensAlways available to the agent
SKILL.md
2,469 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Measurement Tools
- Budgets
- Anti-Patterns
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,266 bytesSKILL.mdcore10,266 bytes
Read SKILL.md
# Performance Tuning
Intuition about where time goes is wrong more often than it is right. You measure the problem, profile to find where the time or memory actually goes, change one thing, and measure again with the same method. **No measurement, no optimisation: every change is justified by a number before and proven by a number after.**
## The Prime Directive
```
NO OPTIMISATION WITHOUT A MEASUREMENT PROVING THE PROBLEM
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubproject-bootstrap
When your agent uses it
Use when creating a new project or repository, scaffolding a codebase from nothing, turning a prototype into something that will be kept, or inheriting a project that has no version control hygiene, formatter, linter, tests or CI.
What it loads into context
Frontmatter
56 tokensAlways available to the agent
SKILL.md
2,272 tokensWhen the skill is activated
Reference documents
2,862 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Project Classes
- The Universal Foundation
- Directory Principles
- Dependency Governance
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,009 bytesSKILL.mdcore10,009 bytesSKILL.mdcore10,009 bytesreferences/templates.mddoc10,457 bytes
Read SKILL.md
# Project Bootstrap
A project's foundation is cheapest on day one and most expensive the day after the first feature lands. Before any feature code, the repository gets version control hygiene, a formatter, a linter, a test runner with one passing test, CI that runs them, and safe handling of configuration and secrets. **The foundation is part of the first commit, not a later clean-up.**
## The Prime Directive
```
NO FEATURE CODE UNTIL THE FOUNDATION RUNS GREEN IN CI
```
No exceptions. No workarounds. No shortcuts.
"Runs green" means you ran format check, lint, type check (where the language has one) and tests locally, they passed, and the same commands are in a CI workflow.
View the complete source on GitHubquality-gates
When your agent uses it
Use when setting up or changing a project's lint, type, test, coverage, build, size, audit, complexity or dead-code checks, when a commit, pull request or merge is being prepared, and when a check fails or someone proposes to disable, skip or loosen one.
What it loads into context
Frontmatter
65 tokensAlways available to the agent
SKILL.md
3,014 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- The Eight Gates
- CI Template
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore12,438 bytesSKILL.mdcore12,438 bytesSKILL.mdcore12,438 bytes
Read SKILL.md
# Quality Gates
Anything a machine can check, a machine checks, on every change, in CI, with the same configuration everyone runs locally. A check that only runs when someone remembers is a suggestion. **When a check fails, the code changes; the check does not.**
## The Prime Directive
```
NEVER DISABLE, SKIP OR LOOSEN A CHECK TO MAKE IT PASS
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubrationale
When your agent uses it
Use when the user asks for new work such as a feature, component, integration, service or dependency, especially when the request names a solution instead of a problem, the scope looks larger than the user may expect, or the motivation is that other products have it
What it loads into context
Frontmatter
60 tokensAlways available to the agent
SKILL.md
1,799 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Output Format
- Handling Pushback and Urgency
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore7,673 bytesSKILL.mdcore7,673 bytes
Read SKILL.md
# Rationale
The most expensive feature is the one nobody needed. Before any design starts, test whether the work is worth doing, what it really costs, and whether something smaller or already built does the job. **Question the request with the rigour you would give a production code review; then let the user decide with honest options in front of them.**
## The Prime Directive
```
NO NEW WORK WITHOUT ASKING WHETHER IT IS WORTH DOING
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubreference-first
When your agent uses it
Use when about to design or build an API, database schema, test strategy, CI pipeline, architecture, library integration or any feature that plausibly exists already as open source, a public standard or a well-known product's published design, and no proven reference has been named yet.
What it loads into context
Frontmatter
64 tokensAlways available to the agent
SKILL.md
2,470 tokensWhen the skill is activated
Reference documents
6,610 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Route First
- The Protocol
- Reference Quality Criteria
- Licence Compatibility
- Presenting the Blueprint
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,942 bytesSKILL.mdcore10,942 bytesSKILL.mdcore10,942 bytesSKILL.mdcore10,942 bytesSKILL.mdcore10,942 bytesreferences/api-and-schema-references.mddoc8,980 bytesreferences/architecture-references.mddoc4,435 bytesreferences/github-search.mddoc7,988 bytesreferences/testing-and-ci-references.mddoc6,502 bytes
Read SKILL.md
# Reference First
Most things you are asked to build have been built before by teams who spent months on them and fixed the incidents that followed. This skill routes each task to the right kind of reference, finds two or three good ones, and turns them into a short, cited blueprint before any code is written. **Build on a named, inspected reference, never on what seems plausible.**
## The Prime Directive
```
NO DESIGN WITHOUT A NAMED, INSPECTED REFERENCE
```
No exceptions. No workarounds. No shortcuts.
A reference you remember is not a reference. A reference counts once you have opened it in this session, read the part you are borrowing, and can link to it.
View the complete source on GitHubreview-response
When your agent uses it
Use when code review feedback arrives from the user, a reviewer subagent, another model, a human colleague or an automated tool, before changing any code in response, and especially when the feedback is unclear, looks wrong, or contains several items that may depend on each other.
What it loads into context
Frontmatter
63 tokensAlways available to the agent
SKILL.md
2,158 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Banned Reactions
- How to Acknowledge Valid Feedback
- How to Push Back
- When Your Pushback Was Wrong
- Feedback by Source
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore9,341 bytesSKILL.mdcore9,341 bytes
Read SKILL.md
# Review Response
Review feedback is a set of technical claims about the code. Each claim is checked against the code before anything changes, then fixed, questioned or rejected with evidence. **You owe the reviewer accuracy, not agreement.**
## The Prime Directive
```
EVERY PIECE OF FEEDBACK GETS A TECHNICAL EVALUATION BEFORE ANY CODE CHANGES
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubsecurity-protocol
When your agent uses it
Use when writing or changing code that handles user input, uploaded files, authentication, sessions, authorisation, database queries, shell commands, file paths, outbound requests to user-supplied URLs, HTTP responses and headers, secrets or credentials, deserialisation, or new third-party dependencies.
What it loads into context
Frontmatter
67 tokensAlways available to the agent
SKILL.md
2,459 tokensWhen the skill is activated
Reference documents
1,906 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Input Validation Checklist
- Secrets Management
- Security Headers
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,893 bytesSKILL.mdcore10,893 bytesSKILL.mdcore10,893 bytesreferences/owasp-top-10.mddoc8,699 bytes
Read SKILL.md
# Security Protocol
Every point where data enters your code from outside is a trust boundary: request bodies, headers, cookies, URLs, files, environment, message queues, third-party API responses, and model output. Data that crosses one is hostile until it has been validated for what it claims to be and encoded for where it is going. **Validate on the way in, encode on the way out, authorise every action, and keep secrets out of code.**
## The Prime Directive
```
NO EXTERNAL DATA REACHES A SYSTEM CALL, QUERY OR OUTPUT WITHOUT VALIDATION
```
No exceptions. No workarounds. No shortcuts.
"System call" includes shell commands, file-system paths, outbound HTTP requests, deserialisers and `eval`-like functions. "Query" includes SQL, NoSQL, LDAP, XPath and search engines. "Output" includes HTML, JSON sent to browsers, logs, headers and prompts sent to a model.
View the complete source on GitHubsystem-design
When your agent uses it
Use when choosing or changing architecture, a database, an API style, authentication, caching, queues or background jobs, file and module organisation, hosting, or any external service, or when the user asks what to use, where to deploy, how to scale, or whether to split into services.
What it loads into context
Frontmatter
67 tokensAlways available to the agent
SKILL.md
2,750 tokensWhen the skill is activated
Reference documents
1,315 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Decision Tables
- File Organisation
- Presenting the Recommendation
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore12,595 bytesreferences/hosting-and-services.mddoc5,727 bytes
Read SKILL.md
# System Design
Every component in an architecture is code to maintain, a bill, and a way to fail. This skill picks the simplest structure that meets a requirement the project has now, prefers what the user already runs over anything new, and records why. **Complexity must be paid for by a current, named requirement; "we might need it" pays for nothing.**
## The Prime Directive
```
NO STRUCTURAL COMPLEXITY WITHOUT A CURRENT, NAMED REQUIREMENT
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubtask-planning
When your agent uses it
Use when an approved design, specification or clear requirement exists for work that touches more than one file or behaviour, before any implementation code is written
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
2,332 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- Entry Check
- The Protocol
- Plan Header
- Task Template
- Self-Review Checklist
- Handoff
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,041 bytesSKILL.mdcore10,041 bytes
Read SKILL.md
# Task Planning
Write the implementation plan for a builder who has never seen this codebase or the conversation. They write good code once they know the exact interface and the exact test; what they cannot know is what was decided. **Put every decision in the plan: which files, which names and signatures, which values, which test proves each task, and which command shows it passed.**
## The Prime Directive
```
NO IMPLEMENTATION WITHOUT A SAVED PLAN OF ATOMIC, VERIFIABLE TASKS
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubteam-orchestration
When your agent uses it
Use when a task needs several agents that must coordinate with each other while they work, such as negotiating an interface across layers, debating competing hypotheses, or redirecting each other's research, and the harness supports agent teams.
What it loads into context
Frontmatter
53 tokensAlways available to the agent
SKILL.md
2,194 tokensWhen the skill is activated
Reference documents
1,604 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Entry Protocol
- The Protocol
- Known Limits (Claude Code, 2026-10-25)
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,056 bytesreferences/team-patterns.mddoc7,424 bytes
Read SKILL.md
# Team Orchestration
An agent team is a lead plus several teammates, each its own full session with its own context, sharing a task list and able to message each other directly. That buys mid-work coordination that subagents cannot give: a subagent reports back to its caller and stops, a teammate can tell another teammate "the response shape changed" while both are still working.
**Core principle: form a team only when teammates must talk to each other mid-work.** Teams cost far more tokens than subagents because every teammate is a separate instance, and token use grows with every active teammate. If no teammate ever needs another's discovery before finishing, use `godmode:parallel-execution` or `godmode:delegated-execution`.
<!-- Facts in this skill: https://code.claude.com/docs/en/agent-teams (fetched 2026-10-25) and docs/research/2026-10-25-claude-code-plugin-spec.md section 6. Harness-specific actions and settings: skills/using-godmode/references/<harness>-tools.md. -->
## The Prime Directive
```
NO TEAM WITHOUT AT LEAST TWO PAIRS OF TEAMMATES THAT MUST SHARE FINDINGS MID-WORK
```
View the complete source on GitHubtest-first
When your agent uses it
Use when implementing a feature, fixing a bug, refactoring, or changing behaviour in any codebase, before writing the production code that does it.
What it loads into context
Frontmatter
38 tokensAlways available to the agent
SKILL.md
2,553 tokensWhen the skill is activated
Reference documents
1,501 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Bug Fixes
- Writing Good Tests
- Worked Example
- When You Are Stuck
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,809 bytesreferences/testing-anti-patterns.mddoc6,648 bytes
Read SKILL.md
# Test-First Development
Every change in behaviour starts as a test that fails for the right reason. You write the test, run it, watch it fail, write the least code that makes it pass, and only then clean up. **A test you never saw fail is a test you have no evidence for.**
## The Prime Directive
```
NO PRODUCTION CODE WITHOUT A FAILING TEST THAT YOU WATCHED FAIL
```
No exceptions. No workarounds. No shortcuts.
Production code written before its test is deleted, not kept "for reference" and not adapted while the test is written. Delete it and build it again from the test.
View the complete source on GitHubui-engineering
When your agent uses it
Use when writing or changing frontend component code, page layouts, forms, client state, responsive behaviour, animation or styling, or when working in a codebase that uses a design system or component library such as shadcn/ui, Material UI, Ant Design, Chakra UI or Mantine.
What it loads into context
Frontmatter
67 tokensAlways available to the agent
SKILL.md
2,898 tokensWhen the skill is activated
Reference documents
4,064 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- No Design System Yet
- Composition Over Configuration
- Layout
- State Ownership
- Responsive
- Accessibility Checklist
- Forms
- Motion
- Performance
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore12,735 bytesSKILL.mdcore12,735 bytesreferences/component-patterns.mddoc8,410 bytesreferences/design-systems.mddoc7,571 bytes
Read SKILL.md
# UI Engineering
`godmode:ui-research` decides what to build: the layout reference, the tokens, the pattern and states of each component. This skill builds it inside the project's design system with sound structure: semantic markup, composed components, clear state ownership, every state implemented, accessible by keyboard and screen reader, and fast. **Use what the design system provides, define structure, states and accessibility before markup, and prove the result in a real browser.**
## The Prime Directive
```
NEVER REBUILD WHAT THE DESIGN SYSTEM PROVIDES; NEVER SHIP A COMPONENT WITHOUT ITS STATES AND KEYBOARD PATH
```
No exceptions. No workarounds. No shortcuts.
## When to Use
View the complete source on GitHubui-research
When your agent uses it
Use when about to design or build any user interface: a website, landing page, web or mobile app screen, dashboard, admin panel, form, game menu or HUD, or a single new component, and no layout reference, design tokens or established design system has been identified for it yet.
What it loads into context
Frontmatter
68 tokensAlways available to the agent
SKILL.md
2,285 tokensWhen the skill is activated
Reference documents
6,701 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Presenting Layout Options
- Hallmarks of Amateur UI
- Component Validation Checklist
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore10,166 bytesreferences/design-tokens.mddoc7,783 bytesreferences/pattern-libraries.mddoc13,438 bytesreferences/template-marketplaces.mddoc6,706 bytes
Read SKILL.md
# UI Research
Interfaces generated from nothing look generated: random spacing, a palette of accidents, every section centred, no empty or error states. Professional templates and documented design systems already encode the layout, hierarchy and states that work. This skill finds those references, fixes the design tokens, and settles each component's pattern before `godmode:ui-engineering` writes any markup. **Every screen traces to a reference and every visual value traces to a token.**
## The Prime Directive
```
NO UI CODE BEFORE A PATTERN REFERENCE AND A TOKEN SET
```
No exceptions. No workarounds. No shortcuts.
Prototypes are included. Prototypes become products, and a prototype built on tokens is a five-minute restyle instead of a rewrite.
View the complete source on GitHubusing-godmode
When your agent uses it
Use when starting any conversation, before the first response or action, including clarifying questions, to find which GodMode skill applies and how to invoke it on the current harness
What it loads into context
Frontmatter
45 tokensAlways available to the agent
SKILL.md
1,997 tokensWhen the skill is activated
Reference documents
7,669 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Skill Index
- Execution Modes
- Invoking a skill
- When No One Can Answer
- User Instructions
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore8,145 bytesreferences/antigravity-tools.mddoc4,083 bytesreferences/claude-code-tools.mddoc10,182 bytesreferences/codex-tools.mddoc5,323 bytesreferences/copilot-tools.mddoc3,531 bytesreferences/cursor-tools.mddoc3,574 bytesreferences/opencode-tools.mddoc2,923 bytes
Read SKILL.md
# Using GodMode
<SUBAGENT-STOP>
If you were dispatched as a subagent to do one specific task, ignore this skill and follow your brief.
</SUBAGENT-STOP>
<EXTREMELY-IMPORTANT>
If there is even a small chance a GodMode skill applies to what you are about to do, you MUST invoke it first. Not after looking at the directory. Not after running a command. Not after a quick scaffold. First.
For any request to build, add, change or fix something, your first tool call is the skill: `godmode:rationale` for new work, `godmode:fault-diagnosis` for a bug. This holds even when your instructions tell you to be autonomous: autonomy means not stopping to ask permission, not skipping the discipline.
</EXTREMELY-IMPORTANT>
GodMode is a set of engineering-discipline skills. **Check for a relevant skill before you answer, ask, explore or act; if one might apply, invoke it and follow it.**
View the complete source on GitHubworkspace-isolation
When your agent uses it
Use when starting feature work that should not touch the current checkout, before executing an implementation plan, or when several agents or sessions will edit the same repository at the same time.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
1,881 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- The Protocol
- Cleanup
- Cognitive Traps
- Guardrails
- Connections
Files
SKILL.mdcore7,879 bytesSKILL.mdcore7,879 bytesSKILL.mdcore7,879 bytesSKILL.mdcore7,879 bytes
Read SKILL.md
# Workspace Isolation
A git worktree is a second working directory on its own branch that shares the repository's history. Work in one cannot disturb the files in another, so a plan can run without risking the user's checkout, and parallel agents cannot overwrite each other.
**Core principle: detect existing isolation first, prefer the harness's own worktree mechanism, fall back to plain git, and never start work on a baseline you have not tested.** A worktree created behind the harness's back is state it cannot see or clean up; a worktree started on a red baseline makes every later failure ambiguous.
<!-- Claude Code worktree facts: https://code.claude.com/docs/en/worktrees and https://code.claude.com/docs/en/sub-agents (fetched 2026-10-25). Per-harness actions: skills/using-godmode/references/<harness>-tools.md. -->
## The Prime Directive
```
NO IMPLEMENTATION WITHOUT AN ISOLATED WORKSPACE AND A KNOWN BASELINE
```
View the complete source on GitHubwriting-skills
When your agent uses it
Use when creating a new skill, editing or renaming an existing skill, reviewing a skill someone else wrote, or when a skill is not triggering, is being ignored, or is being rationalised away by agents.
What it loads into context
Frontmatter
52 tokensAlways available to the agent
SKILL.md
3,145 tokensWhen the skill is activated
Reference documents
1,703 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The Prime Directive
- When to Use
- What Makes a Skill Worth Writing
- The Protocol
- The GodMode Template
- Description Rules
- Actions, Not Tools
- Facts That Can Change
- Running Evals
- Cognitive Traps
- Guardrails
- Connections
Scripts executed separately
../../bin/godmode../../hooks/run-hook.cmd../../scripts/validate.js../../scripts/validate.js../../scripts/validate.js
Files
../../bin/godmoderuns4,670 bytes../../hooks/run-hook.cmdruns3,118 bytes../../scripts/validate.jsruns17,686 bytes../../scripts/validate.jsruns17,686 bytes../../scripts/validate.jsruns17,686 bytes../../scripts/version-files.jsonasset614 bytesSKILL.mdcore13,798 bytesSKILL.mdcore13,798 bytesSKILL.mdcore13,798 bytesSKILL.mdcore13,798 bytesreferences/testing-skills.mddoc7,140 bytes
Read SKILL.md
# Writing Skills
A skill is code that runs on a model. Like code, it is written against a failing test: you first watch an agent get the task wrong without the skill, then write the smallest skill that fixes what you saw, then attack it until the agent stops finding ways around it. **If you never watched an agent fail without the skill, you do not know what the skill has to prevent.**
## The Prime Directive
```
NO SKILL SHIPS WITHOUT A BASELINE FAILURE AND A VERIFIED PASS
```
No exceptions. No workarounds. No shortcuts.
This applies to edits too. A "small wording fix" to a discipline skill can open a loophole; re-run its scenarios.
View the complete source on GitHubSkills included
- agent-briefing
- code-review
- codebase-conventions
- completion-gate
- cross-model-review
- delegated-execution
- doctor
- environment-awareness
- error-recovery
- fault-diagnosis
- grounding
- inline-execution
- intent-discovery
- knowledge-capture
- merge-protocol
- model-routing
- parallel-execution
- performance-tuning
- project-bootstrap
- quality-gates
- rationale
- reference-first
- review-response
- security-protocol
- system-design
- task-planning
- team-orchestration
- test-first
- ui-engineering
- ui-research
- using-godmode
- workspace-isolation
- writing-skills
Repository facts
- License
- MIT
- Created
- Mar 9, 2026
- Last pushed
- Nov 2, 2026
- Forks
- 32
- Open issues
- 1
- Project activity
- Active
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 4, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 41 notes
skills/code-review/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/error-recovery/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/fault-diagnosis/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/grounding/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/merge-protocol/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/merge-protocol/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/merge-protocol/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/performance-tuning/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/project-bootstrap/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 3 dangerous shell commandsin 3 files: bin/godmode:54 delegated-execution/references/ledger-and-workspace.md:142 environment-awareness/references/platform-commands.md:27
- 371 network referencesin 38 files: ui-research/references/pattern-libraries.md:9 ui-research/references/template-marketplaces.md:18 system-design/references/hosting-and-services.md:18 and 35 more
- 1 obfuscation signalin 1 file: scripts/validate.js:207
- No instruction overrides
- 6 executable scriptsin 4 files: scripts/validate.js:1 bin/godmode:1 hooks/run-hook.cmd:1 and 1 more
- No access to secrets
- No requests to vote in Hot
Show all 381 signals by file
.claude-plugin/marketplace.json
- Network accessline 4
"owner": { "name": "David Innes", "url": "https://github.com/NoobyGains" - Network accessline 5
"name": "David Innes", "url": "https://github.com/NoobyGains" }, - Network accessline 6
"url": "https://github.com/NoobyGains" - Network accessline 15
"author": { "name": "David Innes", "url": "https://github.com/NoobyGains" - Network accessline 16
"name": "David Innes", "url": "https://github.com/NoobyGains" }, - Network accessline 17
"url": "https://github.com/NoobyGains"
.claude-plugin/plugin.json
- Network accessline 6
"author": { "name": "David Innes", "url": "https://github.com/NoobyGains" - Network accessline 7
"name": "David Innes", "url": "https://github.com/NoobyGains" }, - Network accessline 8
"url": "https://github.com/NoobyGains" - Network accessline 9
}, "homepage": "https://github.com/NoobyGains/godmode", "repository": "https://github.com/NoobyGains/godmode", - Network accessline 10
"homepage": "https://github.com/NoobyGains/godmode", - Network accessline 11
"repository": "https://github.com/NoobyGains/godmode", - Network accessline 12
"documentationUrl": "https://github.com/NoobyGains/godmode#readme", - Network accessline 13
"supportUrl": "https://github.com/NoobyGains/godmode/issues",
bin/godmode
hooks/run-hook.cmd
- Executable scriptline 1
Executable file
scripts/validate.js
- Executable scriptline 1
#!/usr/bin/env node - Executable scriptline 1
Executable file; content unavailable - Executable scriptline 1
Executable file; content unavailable - Possible obfuscationline 207
if (exists('skills/using-godmode/SKILL.md') && /[\x00-\x08\x0b\x0c\x0e-\x1f]/.test(read('skills/using-godmode/SKILL.md'))) {
skills/completion-gate/SKILL.md
- Network accessline 4
# Claude Code arms this one-shot stop check when the skill is invoked and removes it # after its first pass; other harnesses ignore the field. # https://code.claude.com/docs/en/hooks#hooks-in-skills-and-agents - Network accessline 5
# after its first pass; other harnesses ignore the field. # https://code.claude.com/docs/en/hooks#hooks-in-skills-and-agents hooks: - Network accessline 6
# https://code.claude.com/docs/en/hooks#hooks-in-skills-and-agents
skills/delegated-execution/references/ledger-and-workspace.md
- Dangerous shell operationline 142
Before deleting anything, copy every `Ruling:` line (in order) and every `minor (deferred)` line into the final message to the user. Delete the plan's directory (`rm -rf "$ws"`) only after the work has landed or the user chose to keep the b
skills/environment-awareness/SKILL.md
- Network accessline 53
<!-- Verified 2026-10-25: bun.lock is the text default since Bun 1.2 (https://bun.sh/docs/install/lockfile); uv.lock and .venv (https://docs.astral.sh/uv/guides/projects/). --> - Network accessline 72
<!-- Corepack is not distributed with Node.js from v25 (https://nodejs.org/dist/latest/docs/api/corepack.html). --> - Network accessline 79
1. **Three shells, three syntaxes.** cmd, PowerShell (Windows PowerShell 5.1 and PowerShell 7) and Git Bash (an MSYS2 bash) can all exist on one machine. Quoting, variables, redirection and chaining differ; the mapping table in the referenc - Network accessline 80
<!-- https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_pipeline_chain_operators --> - Network accessline 82
2. **The `bash` on PATH may be the WSL launcher.** Windows can carry a `bash.exe` that starts WSL (in `C:\Windows\System32` or under `WindowsApps`), and it can resolve before Git Bash. Calling it enters a different Linux environment without - Network accessline 83
<!-- https://claudeissues.com/issue/23556-bug-windows-hook-sh-auto-detection-resolves-to-wsl-bash-exe-instead-of-git-bash ; https://github.com/gitpython-developers/GitPython/pull/2201 --> - Network accessline 85
3. **Paths have three forms.** `C:\Users\me` (Windows), `C:/Users/me` (mixed, accepted by most Windows programs and by Node, Python and Git) and `/c/Users/me` (Git Bash). Convert inside Git Bash with `cygpath -w`, `cygpath -m` or `cygpath - - Network accessline 86
<!-- https://www.msys2.org/docs/filesystem-paths/ --> - Network accessline 88
4. **Git Bash rewrites arguments that start with `/`.** Passing `/opt/x`, or a flag such as `-L/regex/`, to a native Windows program turns it into a Windows path. Prefix the command with `MSYS_NO_PATHCONV=1`, or double the leading slash, wh - Network accessline 89
<!-- https://github.com/git-for-windows/build-extra/blob/main/ReleaseNotes.md (Known issues) --> - Network accessline 91
5. **`python` and `python3` may be Store aliases.** On Windows these names can be app execution aliases that open the Microsoft Store instead of running Python. Use the `py` launcher or the project venv's `Scripts\python.exe`, and confirm w - Network accessline 92
<!-- https://docs.python.org/3/using/windows.html --> - Network accessline 94
Also: virtual environments keep executables in `Scripts\` on Windows and `bin/` elsewhere; PowerShell may refuse to run `Activate.ps1` until the user allows local scripts; and activation is rarely needed, because calling the venv's interpre - Network accessline 95
<!-- https://docs.python.org/3/library/venv.html -->
skills/environment-awareness/references/platform-commands.md
- Dangerous shell operationline 27
| Delete file / tree | `rm f` / `rm -rf d` | `Remove-Item f` / `Remove-Item -Recurse -Force d` | `del f` / `rmdir /s /q d` | - Network accessline 81
| Databases | `psql --version`, `mysql --version`, `sqlite3 --version`, `mongosh --version`, `redis-cli --version` | | Cloud and hosting | `aws --version`, `gcloud --version`, `az --version`, `vercel --version`, `fly version`, `netlify --ve - Network accessline 82
| Cloud and hosting | `aws --version`, `gcloud --version`, `az --version`, `vercel --version`, `fly version`, `netlify --version` | | Data and media | `jq --version`, `curl --version`, `ffmpeg -version` | | Runtimes | `node --version`, `pyt - Network accessline 83
| Data and media | `jq --version`, `curl --version`, `ffmpeg -version` | - Network accessline 90
- PowerShell pipeline chain operators (`&&`, `||` from PowerShell 7): https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_pipeline_chain_operators - Network accessline 91
- MSYS2 path forms and `cygpath`: https://www.msys2.org/docs/filesystem-paths/ - Network accessline 92
- Git for Windows path conversion and `MSYS_NO_PATHCONV`: https://github.com/git-for-windows/build-extra/blob/main/ReleaseNotes.md - Network accessline 93
- Python on Windows (`py` launcher, Store aliases): https://docs.python.org/3/using/windows.html - Network accessline 94
- Python `venv` (activation per shell, `VIRTUAL_ENV`): https://docs.python.org/3/library/venv.html - Network accessline 95
- WSL `bash.exe` resolving before Git Bash (field reports): https://claudeissues.com/issue/23556-bug-windows-hook-sh-auto-detection-resolves-to-wsl-bash-exe-instead-of-git-bash and https://github.com/gitpython-developers/GitPython/pull/2201
skills/fault-diagnosis/references/condition-based-waiting-example.ts
- Executable scriptline 1
Executable file
skills/grounding/SKILL.md
- Network accessline 84
<!-- source: https://example.com/docs/config#timeout (fetched 2026-10-25, v4.2) -->
skills/intent-discovery/references/builder-check-prompt.md
- Network accessline 34
<!-- Pattern adapted from obra/superpowers 7.0.1, skills/brainstorming/builder-check-prompt.md (MIT), https://github.com/obra/superpowers/blob/main/skills/brainstorming/builder-check-prompt.md, read 2026-10-25. --> - Network accessline 35
https://github.com/obra/superpowers/blob/main/skills/brainstorming/builder-check-prompt.md, read 2026-10-25. -->
skills/model-routing/SKILL.md
- Network accessline 12
<!-- Model order facts: https://code.claude.com/docs/en/sub-agents and https://code.claude.com/docs/en/agent-teams (fetched 2026-10-25): an omitted per-dispatch model falls back to the agent definition, then a configured default, then the m
skills/performance-tuning/SKILL.md
- Network accessline 73
<!-- Commands checked 2026-10-25 against: https://github.com/sharkdp/hyperfine ; https://github.com/benfred/py-spy ; https://nodejs.org/api/cli.html ; https://pkg.go.dev/cmd/go ; https://www.postgresql.org/docs/current/sql-explain.html ; ht - Network accessline 91
<!-- Core Web Vitals thresholds from https://web.dev/articles/vitals , fetched 2026-10-25 -->
skills/project-bootstrap/references/templates.md
- Network accessline 201
- Release versions: GitHub releases API for eslint/eslint, prettier/prettier, biomejs/biome, microsoft/TypeScript, vitest-dev/vitest, pnpm/pnpm, oven-sh/bun, astral-sh/uv, astral-sh/ruff, golangci/golangci-lint, actions/checkout, actions/se - Network accessline 202
- ESLint config files and init command: https://eslint.org/docs/latest/use/configure/configuration-files and https://eslint.org/docs/latest/use/getting-started - Network accessline 203
- Prettier config files: https://prettier.io/docs/configuration - Network accessline 204
- Vitest config lookup: https://vitest.dev/config/ - Network accessline 205
- pytest config files: https://docs.pytest.org/en/stable/reference/customize.html - Network accessline 206
- Ruff configuration: https://docs.astral.sh/ruff/configuration/ - Network accessline 207
- uv projects: https://docs.astral.sh/uv/guides/projects/ - Network accessline 208
- Bun lockfile and frozen installs: https://bun.sh/docs/install/lockfile and https://bun.sh/docs/cli/install - Network accessline 209
- Biome CLI: https://biomejs.dev/reference/cli/ - Network accessline 210
- pnpm/action-setup inputs: https://github.com/pnpm/action-setup - Network accessline 211
- golangci-lint v2 config and commands: https://golangci-lint.run/docs/configuration/file/ and https://golangci-lint.run/docs/welcome/quick-start/ - Network accessline 212
- Clippy in CI: https://doc.rust-lang.org/clippy/usage.html - Network accessline 213
- rustfmt `--check`: https://github.com/rust-lang/rustfmt - Network accessline 214
- govulncheck: https://go.dev/doc/tutorial/govulncheck - Network accessline 215
- GitHub Actions secure use (SHA pinning, token permissions): https://docs.github.com/en/actions/reference/security/secure-use - Network accessline 216
- TypeScript 7 default changes (secondary reports, not official notes): https://github.com/microsoft/TypeScript/issues/62333
skills/quality-gates/SKILL.md
- Network accessline 62
<!-- ESLint CLI: --max-warnings fails the run when warnings exceed the number; --report-unused-disable-directives. https://eslint.org/docs/latest/use/command-line-interface (fetched 2026-10-25) --> - Network accessline 70
<!-- TSConfig reference: strict, noUncheckedIndexedAccess, noImplicitReturns, noFallthroughCasesInSwitch, noImplicitOverride, exactOptionalPropertyTypes, noEmit. https://www.typescriptlang.org/tsconfig/ (fetched 2026-10-25) --> - Network accessline 73
<!-- mypy --strict: https://mypy.readthedocs.io/en/stable/command_line.html (fetched 2026-10-25) --> - Network accessline 82
<!-- Jest coverageThreshold (global branches/functions/lines/statements; run fails if not met): https://jestjs.io/docs/configuration ; Vitest coverage.thresholds incl. autoUpdate: https://vitest.dev/config/coverage ; pytest-cov --cov-fail-u - Network accessline 95
<!-- size-limit: package.json "size-limit" array with path and limit; throws when over limit. https://github.com/ai/size-limit (fetched 2026-10-25) --> - Network accessline 100
<!-- npm audit --audit-level (info|low|moderate|high|critical|none): https://docs.npmjs.com/cli/v11/commands/npm-audit ; pip-audit exit 1 on findings: https://pypi.org/project/pip-audit/ ; govulncheck ./...: https://go.dev/doc/tutorial/govu - Network accessline 113
<!-- ESLint complexity rule, default max 20: https://eslint.org/docs/latest/rules/complexity ; Ruff C901 with lint.mccabe.max-complexity: https://docs.astral.sh/ruff/rules/complex-structure/ (fetched 2026-10-25) --> - Network accessline 118
<!-- knip finds unused files, exports and dependencies: https://knip.dev/ ; vulture finds unused Python code, exit code 3 when found, --min-confidence: https://pypi.org/project/vulture/ (fetched 2026-10-25) --> - Network accessline 128
<!-- Action majors from the GitHub releases API on 2026-10-25: actions/checkout v7.0.1, actions/setup-node v7.1.0; setup-node accepts node-version 'lts/*' and cache 'npm': https://github.com/actions/setup-node -->
skills/reference-first/SKILL.md
- Network accessline 111
1. Read the reference's `LICENSE` file, not only the badge or the registry metadata (registries often say `NOASSERTION` or "see licence file"). 2. Check compatibility with the project's licence using the table in [github-search.md](referenc - Network accessline 112
2. Check compatibility with the project's licence using the table in [github-search.md](references/github-search.md). 3. No licence means default copyright: study it, do not copy it. <!-- https://docs.github.com/en/repositories/managing-you - Network accessline 113
3. No licence means default copyright: study it, do not copy it. <!-- https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/licensing-a-repository -->
skills/reference-first/references/api-and-schema-references.md
- Network accessline 9
| Source | Study it for | URL | |---|---|---| | RFC 9457, Problem Details for HTTP APIs | A standard machine-readable error body; obsoletes RFC 7807 | https://www.rfc-editor.org/rfc/rfc9457 | - Network accessline 10
|---|---|---| | RFC 9457, Problem Details for HTTP APIs | A standard machine-readable error body; obsoletes RFC 7807 | https://www.rfc-editor.org/rfc/rfc9457 | | Google API Improvement Proposals (AIP) | Resource-oriented design (AIP-121), e - Network accessline 11
| RFC 9457, Problem Details for HTTP APIs | A standard machine-readable error body; obsoletes RFC 7807 | https://www.rfc-editor.org/rfc/rfc9457 | - Network accessline 12
| Google API Improvement Proposals (AIP) | Resource-oriented design (AIP-121), errors (AIP-193), pagination (AIP-158), long-running operations (AIP-151) | https://google.aip.dev | - Network accessline 13
| Microsoft REST API Guidelines | Large-organisation REST conventions, Azure service guidance | https://github.com/microsoft/api-guidelines | - Network accessline 14
| Zalando RESTful API and Event Guidelines | API-first, OpenAPI as the contract, naming, versioning, event formats | https://opensource.zalando.com/restful-api-guidelines/ | - Network accessline 15
| Standard Webhooks | Webhook signing, replay protection, interoperable payloads | https://www.standardwebhooks.com | - Network accessline 21
| Concern | What Stripe does | Source | |---|---|---| | Errors | 2xx success, 4xx client errors, rare 5xx; error object with `type`, `code`, `message`, `param` | https://docs.stripe.com/api/errors | - Network accessline 22
|---|---|---| | Errors | 2xx success, 4xx client errors, rare 5xx; error object with `type`, `code`, `message`, `param` | https://docs.stripe.com/api/errors | | Pagination | Cursor pagination with `limit` (1-100, default 10), `starting_afte - Network accessline 23
| Errors | 2xx success, 4xx client errors, rare 5xx; error object with `type`, `code`, `message`, `param` | https://docs.stripe.com/api/errors | - Network accessline 24
| Pagination | Cursor pagination with `limit` (1-100, default 10), `starting_after`, `ending_before` (mutually exclusive); list object with `data` and `has_more`. The `/v2` namespace paginates differently | https://docs.stripe.com/api/pagin - Network accessline 25
| Idempotency | Client-generated idempotency key (V4 UUID suggested, up to 255 characters) on every v1 `POST`; retries return the saved first response; mismatched parameters on reuse are an error; keys may be pruned after 24 hours | https:/ - Network accessline 26
| Versioning | Dated versions with names (current at check time: `2026-09-30.endive`); monthly non-breaking releases, breaking changes twice a year; per-request override with the `Stripe-Version` header | https://docs.stripe.com/api/version - Network accessline 27
| Enums | Open enums can grow without a version bump; clients must keep a default branch | https://docs.stripe.com/api/versioning | - Network accessline 31
| Concern | What GitHub does | Source | |---|---|---| | Versioning | Date-named versions chosen with the `X-GitHub-Api-Version` header; requests without it get `2022-11-28`; old versions supported at least 24 months; `Deprecation` and `Suns - Network accessline 32
|---|---|---| | Versioning | Date-named versions chosen with the `X-GitHub-Api-Version` header; requests without it get `2022-11-28`; old versions supported at least 24 months; `Deprecation` and `Sunset` headers before retirement; `410 Gone - Network accessline 33
| Versioning | Date-named versions chosen with the `X-GitHub-Api-Version` header; requests without it get `2022-11-28`; old versions supported at least 24 months; `Deprecation` and `Sunset` headers before retirement; `410 Gone` after | http - Network accessline 34
| Breaking versus additive | Published lists of what counts as breaking (removing or renaming fields, new required parameters, removing enum values) and additive (new fields, optional parameters, enum values) | same page | | Pagination (RES - Network accessline 35
| Pagination (REST) | `link` header with `prev`, `next`, `first`, `last`; `per_page`; endpoints use `page`, `before`/`after` or `since` | https://docs.github.com/en/rest/using-the-rest-api/using-pagination-in-the-rest-api | - Network accessline 36
| Rate limits | `x-ratelimit-limit`, `x-ratelimit-remaining`, `x-ratelimit-used`, `x-ratelimit-reset` (UTC epoch seconds), `x-ratelimit-resource`; 403 or 429 when exceeded; honour `retry-after`; back off exponentially | https://docs.github. - Network accessline 37
| Pagination (GraphQL) | Connections with `first`/`last` (1-100) plus `after`/`before` cursors; `pageInfo { endCursor startCursor hasNextPage hasPreviousPage }` | https://docs.github.com/en/graphql/guides/using-pagination-in-the-graphql-api - Network accessline 41
AIP-121 builds APIs from named resources and a small set of standard methods, with custom methods only where those do not fit. AIP-193 requires a `google.rpc.Status` with canonical codes and messages a non-expert can act on. AIP-158 says to - Network accessline 70
Sources: https://github.com/calcom/cal.com/blob/main/packages/prisma/schema.prisma · https://github.com/mastodon/mastodon/blob/main/db/schema.rb · https://github.com/medusajs/medusa · https://github.com/getlago/lago-api · https://github.com
skills/reference-first/references/architecture-references.md
- Network accessline 7
| Source | What it says, briefly | URL | |---|---|---| | Martin Fowler, "Monolith First" (2015) | Almost all successful microservice stories started as a monolith that was broken up; systems built as microservices from scratch often ended i - Network accessline 8
|---|---|---| | Martin Fowler, "Monolith First" (2015) | Almost all successful microservice stories started as a monolith that was broken up; systems built as microservices from scratch often ended in serious trouble. Microservices carry a - Network accessline 9
| Martin Fowler, "Monolith First" (2015) | Almost all successful microservice stories started as a monolith that was broken up; systems built as microservices from scratch often ended in serious trouble. Microservices carry a "premium" that - Network accessline 10
| Shopify, "Deconstructing the Monolith" (2019) | One of the largest Rails codebases, worked on by more than a thousand developers, evolved into a **modular monolith** instead of microservices | https://shopify.engineering/deconstructing-mo - Network accessline 11
| Shopify Packwerk | The Ruby gem Shopify uses to enforce package boundaries inside a Rails monolith; the reference for "modules with enforced boundaries, one deployable" | https://github.com/Shopify/packwerk | - Network accessline 12
| Segment, "Goodbye Microservices" (2018) | Segment's destinations grew into many separate services and repositories with drifting shared-library versions; the team consolidated back to one service and repository | https://twilio.com/en-us/ - Network accessline 13
| The Twelve-Factor App | Configuration in the environment, stateless processes, backing services as attached resources, dev/prod parity | https://12factor.net | - Network accessline 17
| Source | Use it for | URL | |---|---|---| | C4 model (Simon Brown) | Hierarchical diagrams: system context, containers, components, code; plus landscape, dynamic and deployment views; notation- and tool-independent | https://c4model.com | - Network accessline 18
|---|---|---| | C4 model (Simon Brown) | Hierarchical diagrams: system context, containers, components, code; plus landscape, dynamic and deployment views; notation- and tool-independent | https://c4model.com | | arc42 | A template for arch - Network accessline 19
| C4 model (Simon Brown) | Hierarchical diagrams: system context, containers, components, code; plus landscape, dynamic and deployment views; notation- and tool-independent | https://c4model.com | - Network accessline 20
| arc42 | A template for architecture documentation | https://arc42.org | - Network accessline 21
| Architectural Decision Records | One record per significant decision with its rationale, trade-offs and consequences; the set forms the decision log | https://adr.github.io | - Network accessline 29
| Provider | URL | |---|---| | AWS Well-Architected Framework | https://docs.aws.amazon.com/wellarchitected/latest/framework/welcome.html | - Network accessline 30
|---|---| | AWS Well-Architected Framework | https://docs.aws.amazon.com/wellarchitected/latest/framework/welcome.html | | Google Cloud Architecture Center | https://cloud.google.com/architecture | - Network accessline 31
| AWS Well-Architected Framework | https://docs.aws.amazon.com/wellarchitected/latest/framework/welcome.html | - Network accessline 32
| Google Cloud Architecture Center | https://cloud.google.com/architecture | - Network accessline 33
| Azure Architecture Center | https://learn.microsoft.com/en-us/azure/architecture/ | - Network accessline 37
| Source | Use it for | URL | |---|---|---| | `docker/awesome-compose` (CC0-1.0) | Small multi-service Compose setups for common stacks, to copy freely | https://github.com/docker/awesome-compose | - Network accessline 39
| `docker/awesome-compose` (CC0-1.0) | Small multi-service Compose setups for common stacks, to copy freely | https://github.com/docker/awesome-compose | - Network accessline 43
| Source | What it says | URL | |---|---|---| | SQLite, "Appropriate Uses For SQLite" | SQLite allows one writer at a time, but writes usually take milliseconds so writers take turns; sites with fewer than 100K hits a day should work fine ( - Network accessline 44
|---|---|---| | SQLite, "Appropriate Uses For SQLite" | SQLite allows one writer at a time, but writes usually take milliseconds so writers take turns; sites with fewer than 100K hits a day should work fine (a conservative figure); write-he - Network accessline 45
| SQLite, "Appropriate Uses For SQLite" | SQLite allows one writer at a time, but writes usually take milliseconds so writers take turns; sites with fewer than 100K hits a day should work fine (a conservative figure); write-heavy or multi-s - Network accessline 46
| PostgreSQL JSON types | `json` and `jsonb` columns for semi-structured data inside a relational database | https://www.postgresql.org/docs/current/datatype-json.html | - Network accessline 47
| PostgreSQL full-text search | Built-in text search before reaching for a search engine | https://www.postgresql.org/docs/current/textsearch.html | - Network accessline 48
| PostgreSQL LISTEN/NOTIFY | Simple in-database notifications | https://www.postgresql.org/docs/current/sql-listen.html |
skills/reference-first/references/github-search.md
- Network accessline 9
| Channel | Use it for | Address | |---|---|---| | GitHub repository search | Whole projects, starters, examples | https://github.com/search?type=repositories | - Network accessline 10
|---|---|---| | GitHub repository search | Whole projects, starters, examples | https://github.com/search?type=repositories | | GitHub code search | How others call an API or shape a file | https://github.com/search?type=code | - Network accessline 11
| GitHub repository search | Whole projects, starters, examples | https://github.com/search?type=repositories | - Network accessline 12
| GitHub code search | How others call an API or shape a file | https://github.com/search?type=code | - Network accessline 13
| GitHub topics | Curated clusters of projects | `https://github.com/topics/<topic>` | - Network accessline 14
| `awesome-<topic>` lists | Community-vetted shortlists | search `awesome <topic>` | | npm | JavaScript and TypeScript packages | https://www.npmjs.com | | PyPI | Python packages | https://pypi.org | - Network accessline 15
| npm | JavaScript and TypeScript packages | https://www.npmjs.com | - Network accessline 16
| PyPI | Python packages | https://pypi.org | - Network accessline 17
| crates.io | Rust crates | https://crates.io | - Network accessline 18
| pkg.go.dev | Go modules | https://pkg.go.dev | - Network accessline 19
| RubyGems | Ruby gems | https://rubygems.org | - Network accessline 20
| deps.dev (Open Source Insights) | Dependency graph, licences, advisories across npm, Go, Maven, PyPI, NuGet, Cargo, RubyGems | https://deps.dev | - Network accessline 21
| OpenSSF Scorecard | Automated security-practice score for a repository | https://scorecard.dev | - Network accessline 38
Source: https://docs.github.com/en/search-github/searching-on-github/searching-for-repositories - Network accessline 56
Source: https://docs.github.com/en/search-github/github-code-search/understanding-github-code-search-syntax - Network accessline 120
|---|---|---|---| | MIT, BSD-2/3-Clause, ISC | Yes, keep the notice | Yes | Permissive | | Apache-2.0 | Yes, keep `LICENSE` and `NOTICE` | Yes for GPLv3; **not** GPLv2 | FSF treats Apache-2.0 as compatible with GPLv3 only <!-- https://www.a - Network accessline 121
| MIT, BSD-2/3-Clause, ISC | Yes, keep the notice | Yes | Permissive | | Apache-2.0 | Yes, keep `LICENSE` and `NOTICE` | Yes for GPLv3; **not** GPLv2 | FSF treats Apache-2.0 as compatible with GPLv3 only <!-- https://www.apache.org/licenses - Network accessline 122
| Apache-2.0 | Yes, keep `LICENSE` and `NOTICE` | Yes for GPLv3; **not** GPLv2 | FSF treats Apache-2.0 as compatible with GPLv3 only <!-- https://www.apache.org/licenses/GPL-compatibility.html --> | - Network accessline 123
| MPL-2.0 | Yes at file level; changes to MPL files stay MPL | Yes | Weak, file-level copyleft <!-- https://choosealicense.com/licenses/mpl-2.0/ --> | - Network accessline 124
| LGPL | Usually as a separately linked library | Yes | Do not copy LGPL source into non-LGPL files <!-- https://choosealicense.com/licenses/lgpl-3.0/ --> | - Network accessline 125
| GPL-2.0 / GPL-3.0 | No (the whole work would have to become GPL) | GPL-3.0 yes; GPL-2.0-only no | Strong copyleft | | AGPL-3.0 | No | Only if you accept the network-use clause | Running modified code as a network service triggers source s - Network accessline 126
| AGPL-3.0 | No | Only if you accept the network-use clause | Running modified code as a network service triggers source sharing <!-- https://choosealicense.com/licenses/agpl-3.0/ --> | - Network accessline 127
| No licence | No | No | Default copyright: all rights reserved <!-- https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/licensing-a-repository --> | - Network accessline 130
Licence summaries: https://choosealicense.com · identifiers: https://spdx.org/licenses/
skills/reference-first/references/testing-and-ci-references.md
- Network accessline 9
| Job | Reference | Why it is the reference | URL | |---|---|---|---| | UI tests that resemble real use | Testing Library | "The more your tests resemble the way your software is used, the more confidence they can give you." Query the DOM, - Network accessline 10
|---|---|---|---| | UI tests that resemble real use | Testing Library | "The more your tests resemble the way your software is used, the more confidence they can give you." Query the DOM, not component instances | https://testing-library.co - Network accessline 11
| UI tests that resemble real use | Testing Library | "The more your tests resemble the way your software is used, the more confidence they can give you." Query the DOM, not component instances | https://testing-library.com/docs/guiding-pri - Network accessline 12
| Unit and integration (JS/TS) | Vitest (5.x) | Fast runner sharing the Vite config | https://vitest.dev | | End-to-end in real browsers | Playwright (`@playwright/test` 1.6x) | Cross-browser automation and test runner; built-in accessibili - Network accessline 13
| End-to-end in real browsers | Playwright (`@playwright/test` 1.6x) | Cross-browser automation and test runner; built-in accessibility testing guide | https://playwright.dev | | Network mocking (JS) | Mock Service Worker (`msw` 3.x) | Desc - Network accessline 14
| Network mocking (JS) | Mock Service Worker (`msw` 3.x) | Describe the network once with Fetch API semantics; the same handlers work in browser and Node | https://mswjs.io | - Network accessline 15
| Real dependencies in tests | Testcontainers | Throwaway databases, brokers and browsers in Docker, for many languages | https://testcontainers.com | - Network accessline 16
| Contract tests between services | Pact | Consumer-driven contracts for HTTP and message integrations, instead of deploying everything to test | https://docs.pact.io | - Network accessline 17
| Property-based (JS/TS) | fast-check (4.x) | Generated inputs and shrinking for invariants | https://fast-check.dev | - Network accessline 18
| Property-based (Python) | Hypothesis | The same idea for Python | https://hypothesis.readthedocs.io | - Network accessline 19
| Automated accessibility checks | axe-core, with `@axe-core/playwright` | Rule engine for WCAG checks inside browser tests | https://github.com/dequelabs/axe-core | - Network accessline 47
| Source | Study it for | URL | |---|---|---| | `actions/starter-workflows` | Official GitHub Actions templates per language and deploy target | https://github.com/actions/starter-workflows | - Network accessline 48
|---|---|---| | `actions/starter-workflows` | Official GitHub Actions templates per language and deploy target | https://github.com/actions/starter-workflows | | GitHub Actions secure use reference | Hardening rules (below) | https://docs.g - Network accessline 49
| `actions/starter-workflows` | Official GitHub Actions templates per language and deploy target | https://github.com/actions/starter-workflows | - Network accessline 50
| GitHub Actions secure use reference | Hardening rules (below) | https://docs.github.com/en/actions/security-for-github-actions/security-guides/security-hardening-for-github-actions | - Network accessline 51
| `actions/setup-node` (v7 at check time) | Node setup with built-in npm, yarn and pnpm dependency caching | https://github.com/actions/setup-node | - Network accessline 52
| OpenSSF Scorecard action | Automated checks of the repository's own security practices | https://scorecard.dev | - Network accessline 53
| Keep a Changelog | Release notes format | https://keepachangelog.com |
skills/review-response/SKILL.md
- Network accessline 64
<!-- GitHub REST: POST /repos/{owner}/{repo}/pulls/{pull_number}/comments/{comment_id}/replies with body. https://docs.github.com/en/rest/pulls/comments (fetched 2026-10-25) -->
skills/security-protocol/SKILL.md
- Network accessline 94
<!-- Values from the OWASP HTTP Headers Cheat Sheet, fetched 2026-10-25: https://cheatsheetseries.owasp.org/cheatsheets/HTTP_Headers_Cheat_Sheet.html -->
skills/security-protocol/references/owasp-top-10.md
- Network accessline 3
The current edition is **OWASP Top 10:2025** (checked 2026-10-25 at https://top10.owasp.org/2025/). Before relying on this file, confirm the edition is still the latest; OWASP revises the list every few years. - Network accessline 70
**Password storage** (OWASP Password Storage Cheat Sheet, fetched 2026-10-25, https://cheatsheetseries.owasp.org/cheatsheets/Password_Storage_Cheat_Sheet.html): - Network accessline 107
- OWASP Top 10:2025 list and introduction: https://top10.owasp.org/2025/ and https://top10.owasp.org/2025/0x00_2025-Introduction/ - Network accessline 108
- Category pages: https://top10.owasp.org/2025/A01_2025-Broken_Access_Control/ (and the A02 to A10 pages linked from the list) - Network accessline 109
- Password Storage Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Password_Storage_Cheat_Sheet.html - Network accessline 110
- HTTP Headers Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/HTTP_Headers_Cheat_Sheet.html
skills/system-design/SKILL.md
- Network accessline 60
Martin Fowler's observation: successful microservice systems almost always started as a monolith; systems built as microservices from scratch often got into serious trouble. Shopify chose a modular monolith over microservices for one of the - Network accessline 65
|---|---| | Relational data, transactions, multiple users (most applications) | **PostgreSQL** (default) | | Single server or embedded, modest write concurrency, simplest operations | SQLite: one writer at a time, but writes take millisecon - Network accessline 66
| Relational data, transactions, multiple users (most applications) | **PostgreSQL** (default) | | Single server or embedded, modest write concurrency, simplest operations | SQLite: one writer at a time, but writes take milliseconds; the SQ - Network accessline 67
| Single server or embedded, modest write concurrency, simplest operations | SQLite: one writer at a time, but writes take milliseconds; the SQLite authors say sites under 100K hits a day should work fine <!-- https://www.sqlite.org/whentou - Network accessline 68
| Some records with variable shape | PostgreSQL `jsonb` columns before a document database <!-- https://www.postgresql.org/docs/current/datatype-json.html --> | - Network accessline 69
| Text search over application data | PostgreSQL full-text search before a search engine <!-- https://www.postgresql.org/docs/current/textsearch.html --> | - Network accessline 81
| Many client shapes over nested data, client-controlled queries | GraphQL | | Internal service-to-service calls with measured throughput needs, streaming | gRPC | | Server pushes to the browser | Server-sent events for one-way streams; Web - Network accessline 82
| Internal service-to-service calls with measured throughput needs, streaming | gRPC | | Server pushes to the browser | Server-sent events for one-way streams; WebSockets for two-way <!-- https://developer.mozilla.org/en-US/docs/Web/API/Ser - Network accessline 83
| Server pushes to the browser | Server-sent events for one-way streams; WebSockets for two-way <!-- https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events --> | - Network accessline 90
| Context | Choose | |---|---| | A web app with its own users on one site | **Server-side sessions in secure, HttpOnly cookies** (default), via the framework's established auth library <!-- https://cheatsheetseries.owasp.org/cheatsheets/Ses - Network accessline 91
|---|---| | A web app with its own users on one site | **Server-side sessions in secure, HttpOnly cookies** (default), via the framework's established auth library <!-- https://cheatsheetseries.owasp.org/cheatsheets/Session_Management_Cheat - Network accessline 92
| A web app with its own users on one site | **Server-side sessions in secure, HttpOnly cookies** (default), via the framework's established auth library <!-- https://cheatsheetseries.owasp.org/cheatsheets/Session_Management_Cheat_Sheet.htm - Network accessline 97
Never write password hashing, session handling or token validation from scratch. Stateless tokens cannot be revoked before expiry without extra machinery; sessions can. Security review goes to `godmode:security-protocol` (and the OWASP Auth - Network accessline 115
|---|---| | Work fits in the request | **Direct call** (default) | | Occasional slow work, retries, already on PostgreSQL | A Postgres-backed job queue (for example pg-boss for Node.js) <!-- https://github.com/timgit/pg-boss --> | - Network accessline 116
| Work fits in the request | **Direct call** (default) | | Occasional slow work, retries, already on PostgreSQL | A Postgres-backed job queue (for example pg-boss for Node.js) <!-- https://github.com/timgit/pg-boss --> | | Many jobs, Redis - Network accessline 117
| Occasional slow work, retries, already on PostgreSQL | A Postgres-backed job queue (for example pg-boss for Node.js) <!-- https://github.com/timgit/pg-boss --> | - Network accessline 118
| Many jobs, Redis already present | A Redis-backed queue (BullMQ for Node.js, Celery for Python) <!-- https://docs.bullmq.io https://docs.celeryq.dev --> | - Network accessline 119
| Multi-step workflows that must resume after failure, serverless hosting | A durable-workflow service (for example Inngest, which retries from the failed step) <!-- https://www.inngest.com/docs --> |
skills/system-design/references/hosting-and-services.md
- Network accessline 18
| Option | What the vendor says it is | Fits | Check before choosing | Docs | |---|---|---|---|---| | Vercel | Hosting for front-end frameworks with Vercel Functions (Fluid compute) | Next.js and similar front ends, preview deploys | Functi - Network accessline 19
|---|---|---|---|---| | Vercel | Hosting for front-end frameworks with Vercel Functions (Fluid compute) | Next.js and similar front ends, preview deploys | Function limits page; pricing | https://vercel.com/docs/functions/limitations · http - Network accessline 20
| Vercel | Hosting for front-end frameworks with Vercel Functions (Fluid compute) | Next.js and similar front ends, preview deploys | Function limits page; pricing | https://vercel.com/docs/functions/limitations · https://vercel.com/pricing - Network accessline 21
| Netlify | Static and front-end hosting with functions | Static and Jamstack sites | Pricing and build limits | https://www.netlify.com/pricing/ | - Network accessline 22
| Cloudflare Workers | Serverless functions on Cloudflare's network | Edge logic, APIs, static assets, users already on Cloudflare | The limits page (for example CPU time and memory per plan) | https://developers.cloudflare.com/workers/plat - Network accessline 23
| Fly.io | Fly Machines, volumes, networking and managed Postgres; deploy with `fly launch` | Containers close to users, long-lived processes | Pricing, regions, volume backups | https://fly.io/docs/ | - Network accessline 24
| Railway | "All-in-one" cloud for provisioning infrastructure and deploying | Small teams wanting app plus database in one place | Usage-based pricing | https://docs.railway.com | - Network accessline 25
| Render | Cloud platform from prototypes to multi-service architectures | Web services, workers, cron jobs, databases | Instance types and pricing | https://render.com/docs | - Network accessline 26
| VPS (Hetzner Cloud, DigitalOcean and others) | Virtual servers | Steady traffic, Docker workloads, full control, lowest unit cost | The user's willingness to run patching, backups, TLS and monitoring | https://www.hetzner.com/cloud · http - Network accessline 31
| Option | What the vendor says it is | Fits | Docs | |---|---|---|---| | Supabase | Postgres-based backend with framework quickstarts and a CLI | Projects wanting database, auth and storage from one provider | https://supabase.com/docs | - Network accessline 32
|---|---|---|---| | Supabase | Postgres-based backend with framework quickstarts and a CLI | Projects wanting database, auth and storage from one provider | https://supabase.com/docs | | Neon | "The backend for apps and agents": Postgres wi - Network accessline 33
| Supabase | Postgres-based backend with framework quickstarts and a CLI | Projects wanting database, auth and storage from one provider | https://supabase.com/docs | - Network accessline 34
| Neon | "The backend for apps and agents": Postgres with auth, object storage and functions | Serverless Postgres, branch-per-preview workflows | https://neon.com/docs/introduction | - Network accessline 35
| Host-provided Postgres | Managed Postgres from the chosen host (for example Fly.io, Render, Railway) | Keeping the database next to the app on one bill | the host's docs | | Self-managed PostgreSQL | PostgreSQL on the user's own server | - Network accessline 36
| Self-managed PostgreSQL | PostgreSQL on the user's own server | A VPS deployment; the user already runs Postgres | https://www.postgresql.org/docs/current/ | - Network accessline 37
| SQLite | An embedded database file | Single-server apps with modest write concurrency | https://www.sqlite.org/whentouse.html | - Network accessline 41
| Option | Note | Docs | |---|---|---| | Cloudflare R2 | S3-compatible; R2's pricing page says there are no charges for egress bandwidth on any storage class (operations and storage are billed) | https://developers.cloudflare.com/r2/pricing - Network accessline 42
|---|---|---| | Cloudflare R2 | S3-compatible; R2's pricing page says there are no charges for egress bandwidth on any storage class (operations and storage are billed) | https://developers.cloudflare.com/r2/pricing/ | | AWS S3 | The refere - Network accessline 43
| Cloudflare R2 | S3-compatible; R2's pricing page says there are no charges for egress bandwidth on any storage class (operations and storage are billed) | https://developers.cloudflare.com/r2/pricing/ | - Network accessline 44
| AWS S3 | The reference object store; check egress costs for the expected traffic | https://aws.amazon.com/s3/pricing/ | - Network accessline 50
| Option | What it is | Docs | |---|---|---| | pg-boss | Job queue in PostgreSQL for Node.js (MIT) | https://github.com/timgit/pg-boss | - Network accessline 51
|---|---|---| | pg-boss | Job queue in PostgreSQL for Node.js (MIT) | https://github.com/timgit/pg-boss | | BullMQ | Redis-backed job queue for Node.js (MIT) | https://docs.bullmq.io | - Network accessline 52
| pg-boss | Job queue in PostgreSQL for Node.js (MIT) | https://github.com/timgit/pg-boss | - Network accessline 53
| BullMQ | Redis-backed job queue for Node.js (MIT) | https://docs.bullmq.io | - Network accessline 54
| Celery | Distributed task queue for Python | https://docs.celeryq.dev | - Network accessline 55
| Inngest | Durable workflows: runs each step, saves its result, retries from the failed step | https://www.inngest.com/docs | - Network accessline 56
| RabbitMQ | Message broker with flexible routing | https://www.rabbitmq.com/docs | - Network accessline 57
| Apache Kafka | Distributed event log for high-volume streaming and replay | https://kafka.apache.org/documentation/ | - Network accessline 58
| Amazon SQS | Managed message queue on AWS | https://aws.amazon.com/sqs/ |
skills/team-orchestration/SKILL.md
- Network accessline 12
<!-- Facts in this skill: https://code.claude.com/docs/en/agent-teams (fetched 2026-10-25) and docs/research/2026-10-25-claude-code-plugin-spec.md section 6. Harness-specific actions and settings: skills/using-godmode/references/<harness>-t
skills/team-orchestration/references/team-patterns.md
- Network accessline 14
Sizes follow the agent-teams guidance (start with 3 to 5 teammates, 5 to 6 tasks each; https://code.claude.com/docs/en/agent-teams, fetched 2026-10-25).
skills/ui-engineering/SKILL.md
- Network accessline 111
| Rows and columns together (page shell, dashboard grid, aligned card grid) | CSS Grid; `repeat(auto-fill, minmax(<min>, 1fr))` for self-adjusting grids | | One axis (nav items, toolbar, input plus button) | Flexbox with `gap` | | A compone - Network accessline 113
| A component that must adapt to its container, not the viewport | Container queries (Tailwind: `@container` with `@md:` variants) <!-- https://tailwindcss.com/docs/responsive-design --> | - Network accessline 128
Write the small-screen layout first and add at wider breakpoints; Tailwind's breakpoints are mobile-first, so unprefixed utilities apply at every size. <!-- https://tailwindcss.com/docs/responsive-design --> Decide per component what change - Network accessline 132
- [ ] Native elements first: `button` for actions, `a` for navigation, `label` for inputs, landmarks (`header`, `nav`, `main`, `footer`). - [ ] Every interactive element is reachable by Tab, operable by keyboard, and has an accessible name. - Network accessline 133
- [ ] Every interactive element is reachable by Tab, operable by keyboard, and has an accessible name. - [ ] Focus is visible (`:focus-visible`) and never entirely hidden by sticky content (WCAG 2.2 SC 2.4.11). <!-- https://www.w3.org/WAI/W - Network accessline 134
- [ ] Focus is visible (`:focus-visible`) and never entirely hidden by sticky content (WCAG 2.2 SC 2.4.11). <!-- https://www.w3.org/WAI/WCAG22/Understanding/focus-not-obscured-minimum.html --> - Network accessline 135
- [ ] Widgets follow their APG pattern (dialog focus trap and return, tabs with arrow keys, combobox, menu button). <!-- https://www.w3.org/WAI/ARIA/apg/patterns/ --> - Network accessline 136
- [ ] Contrast: text 4.5:1, large text 3:1, component boundaries and focus indicators 3:1. - [ ] Pointer targets at least 24 by 24 CSS pixels or adequately spaced (SC 2.5.8). <!-- https://www.w3.org/WAI/WCAG22/Understanding/target-size-mini - Network accessline 137
- [ ] Pointer targets at least 24 by 24 CSS pixels or adequately spaced (SC 2.5.8). <!-- https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html --> - Network accessline 139
- [ ] Status changes (saved, error, results updated) are announced through a live region. - [ ] Colour is never the only signal. - [ ] Motion respects `prefers-reduced-motion`. <!-- https://developer.mozilla.org/en-US/docs/Web/CSS/@media/pr - Network accessline 141
- [ ] Motion respects `prefers-reduced-motion`. <!-- https://developer.mozilla.org/en-US/docs/Web/CSS/@media/prefers-reduced-motion --> - Network accessline 145
Labels above fields, never placeholder-only; group with `fieldset` and `legend`; correct input types and `autocomplete` values. Validate when the user submits, not on blur or while typing; on failure keep their answers, show an error summar - Network accessline 153
Core Web Vitals at the 75th percentile are the target: LCP within 2.5 s, INP of 200 ms or less, CLS of 0.1 or less. <!-- https://web.dev/articles/vitals -->
skills/ui-engineering/references/component-patterns.md
- Network accessline 5
Library APIs used here were checked on 2026-10-25: `cva` from class-variance-authority (https://cva.style/docs/getting-started/variants), the HTML `dialog` element (https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/Elements/dialog - Network accessline 82
- Label always visible; placeholder is an example at most. - Error text says what is wrong and how to fix it ("Enter an email address in the format name@example.com"), uses colour plus the word or an icon, and is linked with `aria-described - Network accessline 83
- Error text says what is wrong and how to fix it ("Enter an email address in the format name@example.com"), uses colour plus the word or an icon, and is linked with `aria-describedby`. - Validate on submit; on failure keep the value, show - Network accessline 84
- Validate on submit; on failure keep the value, show an error summary at the top of the form and move focus to it (GOV.UK validation pattern, https://design-system.service.gov.uk/patterns/validation/).
skills/ui-engineering/references/design-systems.md
- Network accessline 31
Source: https://ui.shadcn.com/docs/components-json · https://ui.shadcn.com/docs/theming · https://ui.shadcn.com/docs/cli - Network accessline 42
Source: https://mui.com/material-ui/customization/theming/ · https://mui.com/material-ui/customization/theme-components/ - Network accessline 53
Source: https://ant.design/docs/react/customize-theme - Network accessline 63
Source: https://chakra-ui.com/docs/theming/overview · https://chakra-ui.com/docs/get-started/migration - Network accessline 73
Source: https://mantine.dev/theming/theme-object/ · https://mantine.dev/theming/mantine-provider/ · https://mantine.dev/styles/styles-api/ - Network accessline 83
| Library | URL | Note | |---|---|---| | Radix Primitives | https://www.radix-ui.com/primitives | Accessible, unstyled React primitives | - Network accessline 84
|---|---|---| | Radix Primitives | https://www.radix-ui.com/primitives | Accessible, unstyled React primitives | | Base UI | https://base-ui.com | Unstyled React components | - Network accessline 85
| Radix Primitives | https://www.radix-ui.com/primitives | Accessible, unstyled React primitives | - Network accessline 86
| Base UI | https://base-ui.com | Unstyled React components | - Network accessline 87
| React Aria | https://react-spectrum.adobe.com/react-aria/ | Adobe's accessible hooks and components (Apache-2.0) | - Network accessline 88
| Headless UI | https://headlessui.com | Unstyled components from Tailwind Labs | - Network accessline 94
- Tokens live in `@theme` variables in the CSS entry file (Tailwind v4) or in `tailwind.config` (v3 and earlier). <!-- https://tailwindcss.com/docs/theme -->
skills/ui-research/SKILL.md
- Network accessline 115
- [ ] Behaviour is specified at mobile and desktop widths at least. - [ ] Content is realistic, including the longest and shortest plausible values. - [ ] Text contrast is at least 4.5:1, large text at least 3:1, and UI component boundaries - Network accessline 116
- [ ] Content is realistic, including the longest and shortest plausible values. - [ ] Text contrast is at least 4.5:1, large text at least 3:1, and UI component boundaries and states at least 3:1 against adjacent colours. <!-- https://www. - Network accessline 117
- [ ] Text contrast is at least 4.5:1, large text at least 3:1, and UI component boundaries and states at least 3:1 against adjacent colours. <!-- https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html https://www.w3.org/WAI/WCAG - Network accessline 118
- [ ] Pointer targets are at least 24 by 24 CSS pixels or adequately spaced (WCAG 2.2 SC 2.5.8). <!-- https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html --> - Network accessline 119
- [ ] Keyboard path and focus order are defined; the WAI-ARIA Authoring Practices pattern is identified for widgets such as dialogs, tabs, menus and comboboxes. <!-- https://www.w3.org/WAI/ARIA/apg/patterns/ -->
skills/ui-research/references/design-tokens.md
- Network accessline 14
| Radius | none, small, medium, large, full | Consistent shapes | | Shadow / elevation | Three or four levels | Elevation means something | | Breakpoints | The project's breakpoints (Tailwind's defaults are 40, 48, 64, 80 and 96 rem, that i - Network accessline 15
| Shadow / elevation | Three or four levels | Elevation means something | | Breakpoints | The project's breakpoints (Tailwind's defaults are 40, 48, 64, 80 and 96 rem, that is 640 to 1536 px) <!-- https://tailwindcss.com/docs/responsive-des - Network accessline 16
| Breakpoints | The project's breakpoints (Tailwind's defaults are 40, 48, 64, 80 and 96 rem, that is 640 to 1536 px) <!-- https://tailwindcss.com/docs/responsive-design --> | Responsive behaviour planned from the start | - Network accessline 26
shadcn/ui is a working example of the semantic tier: surface tokens pair with `-foreground` tokens (`primary` with `primary-foreground`, `muted` with `muted-foreground`), plus `border`, `input`, `ring`, `destructive` and `chart-1` to `chart - Network accessline 30
The DTCG Format Module 2025.10 is a stable, final Community Group report (not a W3C Standard) for exchanging tokens between tools. <!-- https://www.designtokens.org/TR/2025.10/format/ --> - Network accessline 55
Translation tools such as Style Dictionary convert DTCG files into CSS, JavaScript, iOS and Android outputs. <!-- https://styledictionary.com --> - Network accessline 80
**Tailwind CSS v4.** Theme variables declared with `@theme` create utilities: a `--color-*` variable produces `bg-*`, `text-*` and related utilities, `--font-*` font families, `--text-*` font sizes, `--breakpoint-*` responsive variants. Var - Network accessline 106
- Every text/background pair in the semantic tier meets 4.5:1 (3:1 for large text); focus rings and control borders meet 3:1 against adjacent colours. Check both light and dark themes. <!-- https://www.w3.org/WAI/WCAG22/Understanding/contra
skills/ui-research/references/pattern-libraries.md
- Network accessline 9
| Source | Best for | URL | |---|---|---| | WAI-ARIA Authoring Practices Guide (APG) | Keyboard and screen-reader behaviour for widgets: dialog, tabs, menu button, combobox, disclosure, grid, table, breadcrumb, alert, feed | https://www.w3. - Network accessline 10
|---|---|---| | WAI-ARIA Authoring Practices Guide (APG) | Keyboard and screen-reader behaviour for widgets: dialog, tabs, menu button, combobox, disclosure, grid, table, breadcrumb, alert, feed | https://www.w3.org/WAI/ARIA/apg/patterns/ | - Network accessline 11
| WAI-ARIA Authoring Practices Guide (APG) | Keyboard and screen-reader behaviour for widgets: dialog, tabs, menu button, combobox, disclosure, grid, table, breadcrumb, alert, feed | https://www.w3.org/WAI/ARIA/apg/patterns/ | - Network accessline 12
| WCAG 2.2 | The accessibility success criteria themselves | https://www.w3.org/TR/WCAG22/ | - Network accessline 13
| GOV.UK Design System | Forms, question pages, validation and error messages, tables; heavily user-researched | https://design-system.service.gov.uk/patterns/ | - Network accessline 14
| Material Design 3 | Component specs for Android and the web | https://m3.material.io/components | - Network accessline 15
| Apple Human Interface Guidelines | Platform conventions for iOS, iPadOS, macOS, watchOS, visionOS | https://developer.apple.com/design/human-interface-guidelines/patterns | - Network accessline 16
| IBM Carbon | Enterprise and data-dense UI: data tables, empty states, loading | https://carbondesignsystem.com | - Network accessline 17
| Atlassian Design System | Collaboration and product UI components | https://atlassian.design/components | - Network accessline 18
| Shopify Polaris | Commerce admin UI; Polaris React is deprecated in favour of Polaris web components | https://polaris.shopify.com · https://shopify.dev/docs/api/app-home/polaris-web-components | - Network accessline 19
| Nielsen Norman Group | Research-backed usability articles | https://www.nngroup.com/articles/ | - Network accessline 20
| Mobbin | Screenshots and flows from shipped apps and sites | https://mobbin.com | - Network accessline 21
| Laws of UX | Short summaries of perception and decision heuristics (Hick's law, Fitts's law) | https://lawsofux.com | - Network accessline 22
| Command Line Interface Guidelines | Human-first CLI design | https://clig.dev | - Network accessline 28
**Top bar (marketing, simple apps).** Logo left, a handful of primary links, one primary call to action on the right. On small widths, links move into a menu button that follows the APG menu or disclosure pattern. <!-- https://www.w3.org/WA - Network accessline 30
**Breadcrumbs.** For deep hierarchies only; a `nav` with an accessible name, the current page last and marked with `aria-current`. <!-- https://www.w3.org/WAI/ARIA/apg/patterns/breadcrumb/ --> - Network accessline 32
**Tabs.** For switching between peer views of the same object, not for navigation between pages. Arrow keys move between tabs; Tab moves into the panel. <!-- https://www.w3.org/WAI/ARIA/apg/patterns/tabs/ --> - Network accessline 36
**Layout.** Single column, labels above fields, related fields grouped in a `fieldset` with a `legend`, form width constrained to a readable measure. Mark optional fields "(optional)" rather than marking every required field, or follow the - Network accessline 38
**One thing per page for long flows.** GOV.UK's question-page pattern asks one question per page with a back link and a single "Continue" button; it scales well to multi-step flows (wizards, checkouts). <!-- https://design-system.service.go - Network accessline 40
**Validation.** <!-- https://design-system.service.gov.uk/patterns/validation/ --> - Network accessline 41
- Validate when the user submits or tries to continue, not when they leave a field, and not while they are still typing. GOV.UK's research found early validation causes problems, especially for slower typists. - On failure: show the page ag - Network accessline 42
- On failure: show the page again with the user's answers kept, prefix the page title with "Error: ", show an error summary at the top and move focus to it, and show an error message next to each failing field. <!-- https://design-system.se - Network accessline 48
**Sign-in forms.** Use the right `autocomplete` values and input types so password managers and autofill work. <!-- https://web.dev/articles/sign-in-form-best-practices --> - Network accessline 52
**Data table.** Real `<table>` markup with header cells; right-align numbers; sortable columns show their sort state; sticky header for long tables; row actions in a consistent last column or a row menu; pagination or virtualised scrolling - Network accessline 66
**Modal dialog.** Use for a decision that must happen before continuing (confirm destructive action, short focused task). Focus moves into the dialog and is trapped there; Escape closes it; focus returns to the trigger on close; it has an a - Network accessline 68
**Non-modal alternatives.** Prefer a drawer or inline panel for tasks that need context from the page behind. NN/g describes when modal versus non-modal dialogs fit. <!-- https://www.nngroup.com/articles/modal-nonmodal-dialog/ --> - Network accessline 72
**Toast / notification.** Brief, non-blocking confirmation of something the user did; never the only place an error is reported; does not steal focus; stays long enough to read and can be dismissed; important messages use a live region or t - Network accessline 76
**Empty states.** A blank area makes users wonder whether the system is still loading, failed, or was filtered wrongly. Say what would appear here, why it is empty, and offer the direct next action (create the first item, clear filters, imp - Network accessline 80
**Loading.** - Skeleton screens mimic the final layout and are used for full-page or full-region loads; frame-only skeletons (header and background, no content placeholders) are not recommended; animated shimmer can distract and should resp - Network accessline 81
- Skeleton screens mimic the final layout and are used for full-page or full-region loads; frame-only skeletons (header and background, no content placeholders) are not recommended; animated shimmer can distract and should respect reduced-m - Network accessline 82
- For waits longer than about 10 seconds, show a percent-done indicator and an estimate. <!-- https://www.nngroup.com/articles/progress-indicators/ --> - Network accessline 86
**Error pages.** Page not found: say so plainly, offer search and the main destinations. Service problem: apologise, say what to do and whether data was saved; never show stack traces. GOV.UK has separate patterns for "page not found", "the - Network accessline 98
Search box with a visible label or accessible name and a submit control; recent and suggested queries in a combobox that follows the APG combobox pattern; results show the matched terms, the count, sorting and filters; a no-results state th - Network accessline 114
**CLI output.** Follow the Command Line Interface Guidelines: human-readable output by default with a machine-readable option, errors to stderr with a non-zero exit code, help that shows examples, colour only when the output is a terminal a - Network accessline 118
- **Interaction states** for every interactive element: default, hover, focus-visible, active, disabled, and loading where it triggers work. - **Contrast**: text 4.5:1, large text 3:1, UI component boundaries and states 3:1 (WCAG 2.2 SC 1.4 - Network accessline 119
- **Contrast**: text 4.5:1, large text 3:1, UI component boundaries and states 3:1 (WCAG 2.2 SC 1.4.3 and 1.4.11). <!-- https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html https://www.w3.org/WAI/WCAG22/Understanding/non-text-c - Network accessline 120
- **Target size**: at least 24 by 24 CSS pixels, or enough spacing (WCAG 2.2 SC 2.5.8). Platform guidelines may ask for more on touch devices. <!-- https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html --> - Network accessline 121
- **Focus not obscured**: a focused element must not be entirely hidden by sticky headers, footers or banners (SC 2.4.11). <!-- https://www.w3.org/WAI/WCAG22/Understanding/focus-not-obscured-minimum.html --> - Network accessline 122
- **Colour is never the only signal**: pair it with text, icons or patterns. - **Responsive**: design small widths first, then add columns as space allows. - **Choices**: fewer, clearer options decide faster (Hick's law); make frequent targ - Network accessline 124
- **Choices**: fewer, clearer options decide faster (Hick's law); make frequent targets large and close (Fitts's law). <!-- https://lawsofux.com/hicks-law/ https://lawsofux.com/fittss-law/ -->
skills/ui-research/references/template-marketplaces.md
- Network accessline 18
| Source | What it is | URL | |---|---|---| | ThemeForest (Envato) | WordPress themes, HTML and CMS templates, e-commerce templates | https://themeforest.net | - Network accessline 19
|---|---|---| | ThemeForest (Envato) | WordPress themes, HTML and CMS templates, e-commerce templates | https://themeforest.net | | TemplateMonster | Website templates, plugins and graphics | https://www.templatemonster.com | - Network accessline 20
| ThemeForest (Envato) | WordPress themes, HTML and CMS templates, e-commerce templates | https://themeforest.net | - Network accessline 21
| TemplateMonster | Website templates, plugins and graphics | https://www.templatemonster.com | - Network accessline 22
| Webflow templates | Free and premium Webflow site templates (the page says 7000+) by category: portfolio and agency, technology, professional services, retail | https://webflow.com/templates | - Network accessline 23
| Framer templates | Community templates, filterable by price and style, free and paid | https://www.framer.com/marketplace/ | - Network accessline 24
| Shopify Theme Store | Free and paid storefront themes (for example Dawn, free) | https://themes.shopify.com | - Network accessline 25
| Qode Interactive | Niche-specific premium WordPress themes (agency, creative, and many industries) | https://qodeinteractive.com | - Network accessline 29
| Source | What it is | URL | |---|---|---| | Tailwind Plus | Official components (UI Blocks), React and Next.js site templates and a UI kit from the makers of Tailwind CSS; paid | https://tailwindcss.com/plus | - Network accessline 30
|---|---|---| | Tailwind Plus | Official components (UI Blocks), React and Next.js site templates and a UI kit from the makers of Tailwind CSS; paid | https://tailwindcss.com/plus | | shadcn/ui Blocks | Open-source, copy-and-paste React blo - Network accessline 31
| Tailwind Plus | Official components (UI Blocks), React and Next.js site templates and a UI kit from the makers of Tailwind CSS; paid | https://tailwindcss.com/plus | - Network accessline 32
| shadcn/ui Blocks | Open-source, copy-and-paste React blocks (for example a dashboard with sidebar, charts and data table) | https://ui.shadcn.com/blocks | - Network accessline 33
| Cruip | Tailwind CSS landing pages, websites and dashboards in HTML, React, Next.js and Vue | https://cruip.com | - Network accessline 34
| Vercel Templates | Deployable starters and examples, including framework starters and AI agent templates | https://vercel.com/templates | - Network accessline 35
| Next.js Templates | Free and premium Next.js boilerplates, starters and landing pages | https://nextjstemplates.com | - Network accessline 39
| Source | What it is | URL | |---|---|---| | HTML5 UP | Responsive HTML5 and CSS3 site templates, free under Creative Commons Attribution 3.0, which requires crediting HTML5 UP <!-- https://html5up.net/license --> | https://html5up.net | - Network accessline 40
|---|---|---| | HTML5 UP | Responsive HTML5 and CSS3 site templates, free under Creative Commons Attribution 3.0, which requires crediting HTML5 UP <!-- https://html5up.net/license --> | https://html5up.net | | Start Bootstrap | Free, MIT-l - Network accessline 41
| HTML5 UP | Responsive HTML5 and CSS3 site templates, free under Creative Commons Attribution 3.0, which requires crediting HTML5 UP <!-- https://html5up.net/license --> | https://html5up.net | - Network accessline 42
| Start Bootstrap | Free, MIT-licensed Bootstrap themes, templates and snippets, plus premium products | https://startbootstrap.com | - Network accessline 43
| BootstrapMade | Free and premium Bootstrap templates (the page says 175+) | https://bootstrapmade.com | - Network accessline 44
| Colorlib | Free website templates | https://colorlib.com/wp/templates/ | - Network accessline 48
| Source | What it is | URL | |---|---|---| | AdminLTE | Open-source Bootstrap admin dashboard template, plus premium templates | https://adminlte.io | - Network accessline 49
|---|---|---| | AdminLTE | Open-source Bootstrap admin dashboard template, plus premium templates | https://adminlte.io | | CoreUI | Admin templates and a component library for Bootstrap, React, Angular and Vue | https://coreui.io | - Network accessline 50
| AdminLTE | Open-source Bootstrap admin dashboard template, plus premium templates | https://adminlte.io | - Network accessline 51
| CoreUI | Admin templates and a component library for Bootstrap, React, Angular and Vue | https://coreui.io | - Network accessline 52
| shadcn/ui Blocks | Dashboard and sidebar blocks for React projects (see Tier 2) | https://ui.shadcn.com/blocks | - Network accessline 56
| Source | What it is | URL | |---|---|---| | Mobbin | Searchable library of real mobile apps, web apps and websites: screens, flows (onboarding, checkout, subscription) and UI patterns (dialogs, tables, empty states, paywalls) | https://mo - Network accessline 57
|---|---|---| | Mobbin | Searchable library of real mobile apps, web apps and websites: screens, flows (onboarding, checkout, subscription) and UI patterns (dialogs, tables, empty states, paywalls) | https://mobbin.com | | Awwwards | Award- - Network accessline 58
| Mobbin | Searchable library of real mobile apps, web apps and websites: screens, flows (onboarding, checkout, subscription) and UI patterns (dialogs, tables, empty states, paywalls) | https://mobbin.com | - Network accessline 59
| Awwwards | Award-winning live websites | https://www.awwwards.com | - Network accessline 60
| Godly | Curated web and interface design, typography and motion | https://godly.website | - Network accessline 61
| Lapa Ninja | Landing page gallery (the page says 7,300+ since 2015) with full-page screenshots | https://www.lapa.ninja | - Network accessline 62
| Land-book | Curated website gallery, filterable by page type (landing, pricing, about, blog, product) and industry | https://land-book.com | - Network accessline 63
| One Page Love | Curated one-page sites and section examples, since 2008 | https://onepagelove.com | - Network accessline 64
| SaaS Landing Page | SaaS landing page examples by section (pricing, features, testimonials, FAQ) | https://saaslandingpage.com | - Network accessline 65
| Dribbble | Design concepts and shots; many are unbuilt, so check feasibility | https://dribbble.com |
skills/using-godmode/references/antigravity-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-antigravity-plugins.md; https://antigravity.google/docs/skills ; https://antigravity.google/docs/subagents?tab=cli ; - Network accessline 4
https://antigravity.google/docs/skills ; https://antigravity.google/docs/subagents?tab=cli ; - Network accessline 5
https://antigravity.google/docs/rules ; https://antigravity.google/docs/plugins?app=cli ; - Network accessline 32
- Install: `agy plugin install https://github.com/NoobyGains/godmode`; confirm with `agy plugin list`.
skills/using-godmode/references/claude-code-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-claude-code-plugin-spec.md; https://code.claude.com/docs/en/plugins-reference ; https://code.claude.com/docs/en/skills ; - Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-claude-code-plugin-spec.md; https://code.claude.com/docs/en/plugins-reference ; https://code.claude.com/docs/en/skills ; - Network accessline 4
https://code.claude.com/docs/en/plugins-reference ; https://code.claude.com/docs/en/skills ; - Network accessline 4
https://code.claude.com/docs/en/plugins-reference ; https://code.claude.com/docs/en/skills ; - Network accessline 5
https://code.claude.com/docs/en/sub-agents ; https://code.claude.com/docs/en/agent-teams ; - Network accessline 5
https://code.claude.com/docs/en/sub-agents ; https://code.claude.com/docs/en/agent-teams ; - Network accessline 6
https://code.claude.com/docs/en/hooks (skill hooks, Stop decision control, re-read 2026-10-25) ; - Network accessline 6
https://code.claude.com/docs/en/hooks (skill hooks, Stop decision control, re-read 2026-10-25) ; - Network accessline 7
https://code.claude.com/docs/en/memory ; tool names and parameters checked against the live - Network accessline 7
https://code.claude.com/docs/en/memory ; tool names and parameters checked against the live
skills/using-godmode/references/codex-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-other-harnesses.md; https://learn.chatgpt.com/docs/build-skills ; https://learn.chatgpt.com/docs/agent-configuration/subagents ; - Network accessline 4
https://learn.chatgpt.com/docs/build-skills ; https://learn.chatgpt.com/docs/agent-configuration/subagents ; - Network accessline 5
https://learn.chatgpt.com/docs/agent-configuration/agents-md.md ; https://learn.chatgpt.com/docs/customization/memories ; - Network accessline 38
- Install: `codex plugin marketplace add https://github.com/NoobyGains/godmode`, then `codex plugin add godmode@godmode`.
skills/using-godmode/references/copilot-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-other-harnesses.md; https://docs.github.com/en/copilot/concepts/agents/about-agent-skills ; - Network accessline 4
https://docs.github.com/en/copilot/concepts/agents/about-agent-skills ; - Network accessline 5
https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/create-skills ; - Network accessline 6
https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-custom-agents ; - Network accessline 7
https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/invoke-custom-agents ; - Network accessline 8
https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference ;
skills/using-godmode/references/cursor-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-other-harnesses.md; https://cursor.com/docs/skills ; https://cursor.com/docs/subagents ; https://cursor.com/docs/reference/plugins ; - Network accessline 4
https://cursor.com/docs/skills ; https://cursor.com/docs/subagents ; https://cursor.com/docs/reference/plugins ; - Network accessline 5
https://cursor.com/docs/rules ; https://cursor.com/docs/hooks ;
skills/using-godmode/references/opencode-tools.md
- Network accessline 3
<!-- Sources (fetched 2026-10-25): docs/research/2026-10-25-other-harnesses.md; https://opencode.ai/docs/skills/ ; https://opencode.ai/docs/tools/ ; https://opencode.ai/docs/agents/ ; - Network accessline 4
https://opencode.ai/docs/skills/ ; https://opencode.ai/docs/tools/ ; https://opencode.ai/docs/agents/ ; - Network accessline 5
https://opencode.ai/docs/plugins/ ; https://opencode.ai/docs/rules/ .
skills/workspace-isolation/SKILL.md
- Network accessline 12
<!-- Claude Code worktree facts: https://code.claude.com/docs/en/worktrees and https://code.claude.com/docs/en/sub-agents (fetched 2026-10-25). Per-harness actions: skills/using-godmode/references/<harness>-tools.md. -->
skills/writing-skills/SKILL.md
- Network accessline 138
<!-- Open-standard limits (name 64, description 1,024, body under 500 lines): https://agentskills.io/specification ; Anthropic best practices: https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices (via docs/resear
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Signal confidence: 367 facts
- Executable script in code of a file that runsbin/godmode:1High confidence
- Dangerous shell operation in code of a file that runsbin/godmode:54High confidence
- Executable script in code of a file that runshooks/run-hook.cmd:1High confidence
- Executable script in code of a file that runsscripts/validate.js:1High confidence
- Executable script in code of a file that runsscripts/validate.js:1High confidence
- Executable script in code of a file that runsscripts/validate.js:1High confidence
- Possible obfuscation in code of a file that runsscripts/validate.js:207High confidence
- Network access in the frontmatter of the SKILL.mdcompletion-gate/SKILL.md:4Low confidence
- Network access in the frontmatter of the SKILL.mdcompletion-gate/SKILL.md:5Low confidence
- Network access in the frontmatter of the SKILL.mdcompletion-gate/SKILL.md:6Low confidence
- Dangerous shell operation in prose of documentationdelegated-execution/references/ledger-and-workspace.md:142Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:53Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:72Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:79Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:80Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:82Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:83Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:85Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:86Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:88Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:89Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:91Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:92Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:94Low confidence
- Network access in prose of the SKILL.mdenvironment-awareness/SKILL.md:95Low confidence
- Dangerous shell operation in prose of documentationenvironment-awareness/references/platform-commands.md:27Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:81Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:82Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:83Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:90Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:91Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:92Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:93Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:94Low confidence
- Network access in prose of documentationenvironment-awareness/references/platform-commands.md:95Low confidence
- Executable script in a comment of a file that runsfault-diagnosis/references/condition-based-waiting-example.ts:1Low confidence
- Network access in code of the SKILL.mdgrounding/SKILL.md:84High confidence
- Network access in prose of documentationintent-discovery/references/builder-check-prompt.md:34Low confidence
- Network access in prose of documentationintent-discovery/references/builder-check-prompt.md:35Low confidence
- Network access in prose of the SKILL.mdmodel-routing/SKILL.md:12Low confidence
- Network access in prose of the SKILL.mdperformance-tuning/SKILL.md:73Low confidence
- Network access in prose of the SKILL.mdperformance-tuning/SKILL.md:91Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:201Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:202Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:203Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:204Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:205Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:206Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:207Low confidence
- Network access in prose of documentationproject-bootstrap/references/templates.md:208Low confidence
and 317 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md skills/agent-briefing/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md skills/agent-briefing/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md skills/agent-briefing/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Native | Path .opencode/INSTALL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.