Menu

Not in the top 1,000 by GitHub stars

smixs/skill-conductor

Architecture-first skill lifecycle for AI agents. BinEval binary scoring with threshold-blind, cross-family-calibrated judges, gated self-update loop, pressure testing, 10 authoring principles grounded in empirical research.

Claude CodeTestingProductivityEngineeringResearch

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/skill-conductor/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

3c21d2f19c33 → 3c21d2f19c33

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → 3c21d2f19c33

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit 3c21d2f19c336d3a3333bfe4f45eee041bd13beb

git clone --filter=blob:none --no-checkout https://github.com/smixs/skill-conductor.git .hotskills/smixs-skill-conductor && git -C .hotskills/smixs-skill-conductor checkout --detach 3c21d2f19c336d3a3333bfe4f45eee041bd13beb # Copy each selected skill directory from .hotskills/smixs-skill-conductor into .claude/skills/<skill-name>/
207GitHub stars
6stars in 7 days
1skill in this repository
0commits in 30 days

Heat

3.5
  • 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

Oct 25, 2026Nov 5, 2026207

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

skill-conductor

When your agent uses it

Create, edit, evaluate, and package agent skills. Use when building a new skill from scratch, improving an existing skill, fixing a skill that never triggers or fires unreliably, running evals to test a skill, benchmarking skill performance, optimizing a skill's description, reviewing third-party skills for quality, or packaging skills for distribution — even if the user doesn't explicitly say "skill" (e.g. "teach Claude to do X", "make the agent always follow Y"). Not for using skills or general coding tasks.

What it loads into context

Frontmatter

130 tokens

Always available to the agent

SKILL.md

5,955 tokens

When the skill is activated

Reference documents

31,522 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Runtime requirements (pre-flight)
  2. How to communicate
  3. Modes
  4. Mode 1: CREATE
  5. Mode 2: IMPROVE
  6. Mode 3: VALIDATE
  7. Mode 4: REVIEW
  8. Mode 5: OPTIMIZE
  9. Mode 6: PACKAGE
  10. Quick Reference
  11. Reference Files

Scripts executed separately

  • eval-viewer/generate_review.py
  • scripts/aggregate_benchmark.py
  • scripts/eval_skill.py
  • scripts/generate_report.py
  • scripts/improve_description.py
  • scripts/init_skill.py
  • scripts/package_skill.py
  • scripts/quick_validate.py
  • scripts/run_eval.py
  • scripts/run_loop.py
  • scripts/split_evals.py
  • scripts/test_smoke.py
  • scripts/utils.py

Files

  • ../../.claude-plugin/marketplace.jsonasset486 bytes
  • ../../.claude-plugin/plugin.jsonasset416 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • SKILL.mdcore26,495 bytes
  • agents/analyzer.mddoc16,753 bytes
  • agents/bineval.mddoc12,214 bytes
  • agents/comparator.mddoc8,749 bytes
  • agents/grader.mddoc13,155 bytes
  • assets/eval_review.htmlasset8,502 bytes
  • eval-viewer/generate_review.pyruns16,365 bytes
  • eval-viewer/viewer.htmlasset50,982 bytes
  • references/bineval-method.mddoc9,956 bytes
  • references/patterns.mddoc6,826 bytes
  • references/pressure-testing.mddoc6,258 bytes
  • references/quality-questions.mddoc10,983 bytes
  • references/runtime-setup.mddoc1,839 bytes
  • references/schemas.mddoc21,365 bytes
  • references/sop-practices.mddoc20,562 bytes
  • scripts/aggregate_benchmark.pyruns14,386 bytes
  • scripts/eval_skill.pyruns16,138 bytes
  • scripts/generate_report.pyruns12,847 bytes
  • scripts/improve_description.pyruns11,057 bytes
  • scripts/init_skill.pyruns14,413 bytes
  • scripts/package_skill.pyruns4,263 bytes
  • scripts/quick_validate.pyruns3,874 bytes
  • scripts/run_eval.pyruns11,464 bytes
  • scripts/run_loop.pyruns12,914 bytes
  • scripts/split_evals.pyruns1,949 bytes
  • scripts/test_smoke.pyruns11,691 bytes
  • scripts/utils.pyruns2,941 bytes
Read SKILL.md

# Skill Conductor

Full lifecycle management for agent skills: **draft → test → review → improve → repeat**.

One skill to rule them all — from architecture to packaging. The core loop is always the same: write something, test it, see what fails, fix it, test again.

## Runtime requirements (pre-flight)

Before any mode that touches scripts (CREATE, IMPROVE, VALIDATE, OPTIMIZE, PACKAGE), run the pre-flight block → **`references/runtime-setup.md`** (checks `uv`, sets `UV_BIN`/`SKILL_CONDUCTOR_DIR`, verifies LLM access). If `uv` is absent, stop and tell the user.

## How to communicate

Read context cues. If the user is a skill author iterating on their own work, be direct and technical. If they're new to skills, explain the _why_ behind each step — not just what to do, but why it matters. Default to conversational, not robotic.
View the complete source on GitHub

Skills included

  • skill-conductor

Repository facts

License
MIT
Created
Mar 3, 2026
Last pushed
Aug 3, 2026
Forks
27
Open issues
1
Project activity
Slowing

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 5, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 7 notes
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/skill-conductor/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 63 signals by file

.claude-plugin/plugin.json

  • Network accessline 4
    "description": "Create, edit, evaluate, and package Claude Code agent skills — full draft→test→review→improve lifecycle.",   "author": { "name": "smixs", "email": "shimaozb@gmail.com" },   "homepage": "https://github.com/smixs/skill-conduct
  • Network accessline 5
    "author": { "name": "smixs", "email": "shimaozb@gmail.com" },   "homepage": "https://github.com/smixs/skill-conductor",   "license": "MIT",
  • Network accessline 6
    "homepage": "https://github.com/smixs/skill-conductor",

skills/skill-conductor/assets/eval_review.html

  • Network accessline 5
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />     <title>Eval Set Review - __SKILL_NAME_PLACEHOLDER__</title>     <link rel="preconnect" href="https://fonts.googleapis.com" />
  • Network accessline 6
    <title>Eval Set Review - __SKILL_NAME_PLACEHOLDER__</title>     <link rel="preconnect" href="https://fonts.googleapis.com" />     <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
  • Network accessline 7
    <link rel="preconnect" href="https://fonts.googleapis.com" />
  • Network accessline 8
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
  • Network accessline 9
    <link       href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap"       rel="stylesheet"
  • Network accessline 10
    href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap"

skills/skill-conductor/eval-viewer/generate_review.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Network accessline 449
    url = f"http://localhost:{port}"

skills/skill-conductor/eval-viewer/viewer.html

  • Network accessline 5
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />     <title>Eval Review</title>     <link rel="preconnect" href="https://fonts.googleapis.com" />
  • Network accessline 6
    <title>Eval Review</title>     <link rel="preconnect" href="https://fonts.googleapis.com" />     <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
  • Network accessline 7
    <link rel="preconnect" href="https://fonts.googleapis.com" />
  • Network accessline 8
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
  • Network accessline 9
    <link       href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap"       rel="stylesheet"
  • Network accessline 10
    href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap"
  • Network accessline 12
    />     <script       src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js"
  • Network accessline 13
    <script       src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js"       integrity="sha384-EnyY0/GSHQGSxSgMwaIPzSESbqoOLSexfnSMN2AP+39Ckmn92stwABZynq1JyzdT"
  • Network accessline 14
    src="https://cdn.sheetjs.com/xlsx-0.20.3/package/dist/xlsx.full.min.js"
  • Network accessline 757
    if (!hasPrevious) {           try {             const resp = await fetch("/api/feedback");
  • Network accessline 758
    try {             const resp = await fetch("/api/feedback");             const data = await resp.json();
  • Network accessline 759
    const resp = await fetch("/api/feedback");
  • Network accessline 1144
    fetch("/api/feedback", {
  • Network accessline 1185
    2,         );         fetch("/api/feedback", {
  • Network accessline 1186
    );         fetch("/api/feedback", {           method: "POST",
  • Network accessline 1187
    fetch("/api/feedback", {

skills/skill-conductor/references/pressure-testing.md

  • Network accessline 122
    - obra/superpowers — `writing-skills` skill (micro-test protocol, bulletproofing, rationalization tables) and `testing-skills-with-subagents.md` (pressure types, scenario design) — https://github.com/obra/superpowers

skills/skill-conductor/references/runtime-setup.md

  • Network accessline 6
    # 1. uv (used by every script in scripts/ and eval-viewer/) command -v uv >/dev/null || command -v ~/.local/bin/uv >/dev/null \   || { echo "FAIL: uv not found. Install via curl -LsSf https://astral.sh/uv/install.sh | sh"; exit 1; }
  • Dangerous shell operationline 7
    command -v uv >/dev/null || command -v ~/.local/bin/uv >/dev/null    || { echo "FAIL: uv not found. Install via curl -LsSf https://astral.sh/uv/install.sh | sh"; exit 1; }
  • Network accessline 7
    command -v uv >/dev/null || command -v ~/.local/bin/uv >/dev/null    || { echo "FAIL: uv not found. Install via curl -LsSf https://astral.sh/uv/install.sh | sh"; exit 1; }
  • Dangerous shell operationline 8
    || { echo "FAIL: uv not found. Install via curl -LsSf https://astral.sh/uv/install.sh | sh"; exit 1; }
  • Network accessline 8
    || { echo "FAIL: uv not found. Install via curl -LsSf https://astral.sh/uv/install.sh | sh"; exit 1; }
  • Access to secretsline 26
    [Credential-related source omitted]
  • Access to secretsline 27
    [Credential-related source omitted]

skills/skill-conductor/references/sop-practices.md

  • Access to secretsline 32
    [Credential-related source omitted]
  • Access to secretsline 214
    [Credential-related source omitted]
  • Network accessline 336
    - Penn State Extension — "Standard Operating Procedures: A Writing Guide" (Richard Stup) — https://extension.psu.edu/standard-operating-procedures-a-writing-guide
  • Network accessline 338
    - Toyota TWI (Training Within Industry) — the "Job Instruction" methodology - McDonald's Operations Manual — the canonical example of a 600+ page SOP system for a franchise - anthropics/skills — `skill-creator` skill, the "pushy" descriptio
  • Network accessline 339
    - McDonald's Operations Manual — the canonical example of a 600+ page SOP system for a franchise - anthropics/skills — `skill-creator` skill, the "pushy" description recommendation against undertriggering — https://github.com/anthropics/ski
  • Network accessline 340
    - anthropics/skills — `skill-creator` skill, the "pushy" description recommendation against undertriggering — https://github.com/anthropics/skills
  • Network accessline 341
    - obra/superpowers — `writing-skills` skill, "Match the Form to the Failure" and the wording tests behind it — https://github.com/obra/superpowers
  • Network accessline 342
    - softaworks/agent-toolkit — `skill-judge` skill, the Expert / Activation / Redundant taxonomy and the E:A:R ratio — https://github.com/softaworks/agent-toolkit

skills/skill-conductor/scripts/aggregate_benchmark.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/eval_skill.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/generate_report.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Network accessline 37
    <meta charset="utf-8"> """ + refresh_tag + """    <title>""" + title_prefix + """Skill Description Optimization</title>     <link rel="preconnect" href="https://fonts.googleapis.com">
  • Network accessline 38
    """ + refresh_tag + """    <title>""" + title_prefix + """Skill Description Optimization</title>     <link rel="preconnect" href="https://fonts.googleapis.com">     <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  • Network accessline 39
    <link rel="preconnect" href="https://fonts.googleapis.com">
  • Network accessline 40
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  • Network accessline 41
    <link href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600&family=Lora:wght@400;500&display=swap" rel="stylesheet">

skills/skill-conductor/scripts/improve_description.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/init_skill.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/package_skill.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/quick_validate.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/run_eval.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Access to secretsline 83
    [Credential-related source omitted]

skills/skill-conductor/scripts/run_loop.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/split_evals.py

  • Executable scriptline 1
    #!/usr/bin/env python3

skills/skill-conductor/scripts/test_smoke.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Access to secretsline 141
    [Credential-related source omitted]
  • Dangerous shell operationline 150
    assert result.returncode != 0, "secret leak should fail the eval (non-zero exit)"

skills/skill-conductor/scripts/utils.py

  • Executable scriptline 1
    Executable file

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.

Declared and executed: 2 facts
Signal confidence: 38 facts

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexCompatibleStandard SKILL.md skills/skill-conductor/SKILL.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md skills/skill-conductor/SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md skills/skill-conductor/SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md skills/skill-conductor/SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md skills/skill-conductor/SKILL.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.

You might also use

AmazingAng/old-coder

An old coder's strategy for the agent era: don't read the code — make it run the gauntlet. Evidence-first development skill for coding agents, inspired by Uncle Bob.

757GitHub stars

OneWave-AI/claude-skills

200+ production-ready Claude Code skills for sales, marketing, design, engineering, and AI agent architecture. Built and maintained by OneWave AI.

415GitHub stars

FlorianBruniaux/flow-lean

Model-agnostic output control for LLM agents: concise, action-first answers with proof and risk preserved. Native install paths for Claude Code and Codex.

24GitHub stars

affaan-m/ECC

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

298.3kGitHub stars

DietrichGebert/ponytail

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

185.7kGitHub stars

KKKKhazix/khazix-skills

数字生命卡兹克开源的 AI Skills 合集 | Agent Skills: leader（帮你定义目标）, neat-freak 洁癖, hv-analysis, khazix-writer & more — Claude Code, Codex & 40+ agents

22kGitHub stars