#368 by GitHub stars
chaseai-yt/claudex-loop
Claude Code skill: four-phase plan hardening (recon, interrogate, Codex adversarial review, cross-model build & inspection) — two AI models harden your plan before a line of code exists, then swap jobs to build it. Whoever built it never grades it.
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
legacy/grill-me-codex/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Local scripts
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 53cddc42e393
No earlier version was observed for comparison.
legacy/grill-with-docs-codex/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Local scripts
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 53cddc42e393
No earlier version was observed for comparison.
skills/claudex-loop/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Local scripts → After Local scripts
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Local scripts
First observation → 53cddc42e393
No earlier version was observed for comparison.
skills/claudex-route/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 53cddc42e393
No earlier version was observed for comparison.
skills/codex-build/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 53cddc42e393
No earlier version was observed for comparison.
skills/codex-review/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Instructions only → After Instructions only
53cddc42e393 → 53cddc42e393
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 53cddc42e393
No earlier version was observed for comparison.
Verified commit 53cddc42e39346fb8c6d5d612281a3b1add14e66
git clone --filter=blob:none --no-checkout https://github.com/chaseai-yt/claudex-loop.git .hotskills/chaseai-yt-claudex-loop && git -C .hotskills/chaseai-yt-claudex-loop checkout --detach 53cddc42e39346fb8c6d5d612281a3b1add14e66
# Copy each selected skill directory from .hotskills/chaseai-yt-claudex-loop into .claude/skills/<skill-name>/Heat
7.2- Star growth, last 7 daysAhead of 91% of tracked repositories
- Star growth, last 24 hoursAhead of 90% of tracked repositories
- Rank movement, last 7 daysAhead of 58% of tracked repositories
- Commits, last 30 daysAhead of 80% of tracked repositories
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekAhead of 0% of tracked repositories
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- grill-me-codex
legacy/grill-me-codex - grill-with-docs-codex
legacy/grill-with-docs-codex - claudex-loop
skills/claudex-loop - claudex-route
skills/claudex-route - codex-build
skills/codex-build - codex-review
skills/codex-review
How it works
grill-me-codex
When your agent uses it
SUPERSEDED by /claudex-loop (formerly /crucible; adds a Phase 0 research/recon pass + rebuilt interview) — prefer that skill unless you explicitly want this one. Two-act plan hardening. ACT 1 (you ↔ Claude) — Claude interviews you relentlessly about a plan or design, one question at a time, recommending an answer for each and exploring the codebase when it can answer itself, until every branch of the decision tree is resolved. ACT 2 (Claude ↔ Codex) — Claude writes the locked plan to PLAN.md and OpenAI Codex adversarially reviews it in a read-only sandbox (VERDICT:APPROVED/REVISE), Claude revises and re-submits to the SAME Codex session until APPROVED or a MAX_ROUNDS cap, then you sign off before any code. Use when the user says "/grill-me-codex", "grill me then have codex review", "grill me and stress-test the plan", "interview me about this plan then get a second model on it", or is about to build something high-stakes (auth, schema, concurrency, migrations, payments) and wants both alignment AND a cross-model sanity check before implementation. Builds on Matt Pocock's grill-me (MIT). For the docs-aware variant use /grill-with-docs-codex; if you already have a plan and want only the Codex review use /codex-review. NOT for reviewing already-written code (use /codex:review) and NOT for trivial changes.
What it loads into context
Frontmatter
332 tokensAlways available to the agent
SKILL.md
2,496 tokensWhen the skill is activated
Reference documents
308 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- ACT 1 — GRILL (you ↔ Claude)
- ACT 2 — REVIEW (Claude ↔ Codex)
- Hard rules
- What NOT to do
Files
../../.claude-plugin/marketplace.jsonasset350 bytes../../.claude-plugin/plugin.jsonasset474 bytesSKILL.mdcore9,844 bytesSKILL.mdcore9,844 bytesSKILL.mdcore9,844 bytesTHIRD-PARTY-NOTICES.mddoc1,361 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubgrill-with-docs-codex
When your agent uses it
SUPERSEDED by /claudex-loop (formerly /crucible; its docs-aware mode covers this variant) — prefer that skill unless you explicitly want this one. Two-act plan hardening with living documentation. ACT 1 (you ↔ Claude) — Claude interviews you relentlessly about a plan, one question at a time, challenging it against your project's existing domain model and glossary (CONTEXT.md), sharpening fuzzy terms, stress-testing with concrete scenarios, cross-referencing code, and updating CONTEXT.md + ADRs inline as decisions crystallise. ACT 2 (Claude ↔ Codex) — Claude writes the locked plan to PLAN.md and OpenAI Codex adversarially reviews it in a read-only sandbox (VERDICT:APPROVED/REVISE), Claude revises and re-submits to the SAME Codex session until APPROVED or a MAX_ROUNDS cap, then you sign off before any code. Use when the user says "/grill-with-docs-codex", "grill me against the docs then have codex review", "stress-test this against our domain model then get a second model on it", or is about to build something high-stakes in a project with established terminology/ADRs and wants alignment, documentation, AND a cross-model sanity check. Builds on Matt Pocock's grill-with-docs (MIT). NOT for reviewing already-written code (use /codex:review) and NOT for trivial changes.
What it loads into context
Frontmatter
309 tokensAlways available to the agent
SKILL.md
2,787 tokensWhen the skill is activated
Reference documents
1,504 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- ACT 1 — GRILL WITH DOCS (you ↔ Claude)
- Domain awareness
- During the session
- ACT 2 — REVIEW (Claude ↔ Codex)
- Hard rules
- What NOT to do
Files
ADR-FORMAT.mddoc2,766 bytesCONTEXT-FORMAT.mddoc2,299 bytesSKILL.mdcore11,153 bytesSKILL.mdcore11,153 bytesSKILL.mdcore11,153 bytesTHIRD-PARTY-NOTICES.mddoc1,486 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubclaudex-loop
When your agent uses it
Harden a plan with independent Claude/Codex review, then optionally build and cross-inspect it. Start in either Claude Code or Codex: the host plans and the other provider reviews. Use for claudex this plan, claudex-loop, or the legacy crucible trigger; not for trivial edits.
What it loads into context
Frontmatter
74 tokensAlways available to the agent
SKILL.md
2,026 tokensWhen the skill is activated
Reference documents
3,436 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Resolve roles once
- Tunables
- Phase 0 — Recon
- Phase 1 — Settle requirements
- Phase 2 — Independent plan review
- Phase 3 — Build and inspect
Scripts executed separately
scripts/runner.py
Files
ADR-FORMAT.mddoc1,891 bytesCONTEXT-FORMAT.mddoc1,647 bytesSKILL.mdcore9,951 bytesSKILL.mdcore9,951 bytesreferences/build.mddoc4,987 bytesreferences/runtime.mddoc8,064 bytesscripts/runner.pyruns28,074 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubclaudex-route
When your agent uses it
Recommend a model and a scoped handoff for a task in Claude Code or Codex. Use when choosing who should handle a task, seeking a second opinion, getting unstuck, or delegating focused work; execute one handoff when requested.
What it loads into context
Frontmatter
59 tokensAlways available to the agent
SKILL.md
1,512 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
- Understand the job
- Choose the role before the model
- Select a practical candidate
- Return a short routing brief
- Execute one handoff when requested
Files
SKILL.mdcore7,757 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubcodex-build
When your agent uses it
Have Codex implement a concrete work order and have Claude independently inspect the final changes. Preserves the explicit Codex builder choice; use claudex-loop for host-based planning and automatic reviewer selection.
What it loads into context
Frontmatter
49 tokensAlways available to the agent
SKILL.md
430 tokensWhen the skill is activated
Reference documents
2,640 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../claudex-loop/references/build.mddoc4,987 bytes../claudex-loop/references/runtime.mddoc8,064 bytesSKILL.mdcore2,109 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubcodex-review
When your agent uses it
Have Codex independently review an existing implementation plan while Claude coordinates a bounded revision loop. For automatic reviewer selection from either host, use claudex-loop with mode=review.
What it loads into context
Frontmatter
44 tokensAlways available to the agent
SKILL.md
363 tokensWhen the skill is activated
Reference documents
3,719 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
Files
../claudex-loop/SKILL.mddoc9,951 bytes../claudex-loop/references/runtime.mddoc8,064 bytesSKILL.mdcore1,769 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubSkills included
- grill-me-codex
- grill-with-docs-codex
- claudex-loop
- claudex-route
- codex-build
- codex-review
Repository facts
- License
- No license detected
- Created
- Jun 5, 2026
- Last pushed
- Oct 14, 2026
- Forks
- 358
- Open issues
- 4
- Project activity
- Active
Rank over the last 12 weeks
Now #368 · best #344 · down 2 in 30 days
- #363 on Oct 29
- #364 from Oct 30 to Nov 2
- #365 on Nov 3
- #364 on Nov 4
- #366 on Nov 5
- #368 since Nov 6
77 earlier periods
- #560 on Aug 2
- #529 on Aug 7
- #521 on Aug 8
- #516 on Aug 9
- #529 on Aug 10
- #526 on Aug 11
- #521 on Aug 12
- #520 on Aug 13
- #517 on Aug 14
- #498 on Aug 15
- #506 on Aug 16
- #510 on Aug 17
- #529 on Aug 18
- #519 on Aug 19
- #522 on Aug 21
- #505 on Aug 22
- #529 on Aug 23
- #519 on Aug 24
- #514 on Aug 25
- #498 on Aug 26
- #502 on Aug 27
- #497 on Aug 28
- #484 on Aug 29
- #461 on Aug 30
- #477 from Aug 31 to Sep 1
- #469 on Sep 2
- #472 on Sep 3
- #455 on Sep 4
- #454 on Sep 5
- #460 on Sep 6
- #455 on Sep 7
- #465 on Sep 8
- #452 on Sep 9
- #450 on Sep 10
- #448 on Sep 11
- #446 on Sep 12
- #439 on Sep 13
- #445 on Sep 14
- #427 on Sep 15
- #426 on Sep 16
- #415 on Sep 17
- #407 on Sep 18
- #391 on Sep 19
- #393 on Sep 20
- #385 on Sep 21
- #388 on Sep 22
- #378 on Sep 23
- #376 on Sep 24
- #372 on Sep 25
- #368 from Sep 26 to Sep 27
- #371 on Sep 28
- #378 on Sep 29
- #371 on Sep 30
- #370 on Oct 1
- #371 on Oct 2
- #360 on Oct 3
- #357 on Oct 4
- #363 on Oct 5
- #362 on Oct 6
- #366 on Oct 7
- #367 on Oct 8
- #373 on Oct 9
- #357 on Oct 10
- #353 on Oct 11
- #358 on Oct 12
- #355 on Oct 13
- #357 on Oct 14
- #352 on Oct 15
- #353 on Oct 16
- #345 on Oct 17
- #352 from Oct 18 to Oct 19
- #351 from Oct 20 to Oct 21
- #350 on Oct 22
- #348 on Oct 23
- #344 on Oct 24
- #356 on Oct 25
- #361 on Oct 28
Before you install
Local scriptsAutomated signals, not a guarantee
Checked Nov 2, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 5 notes
legacy/grill-me-codex/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlegacy/grill-me-codex/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlegacy/grill-with-docs-codex/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedlegacy/grill-with-docs-codex/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/claudex-loop/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 17 network referencesin 6 files: .claude-plugin/plugin.json:4 .claude-plugin/marketplace.json:3 claudex-loop/references/runtime.md:57 and 3 more
- No hidden or obfuscated text
- No instruction overrides
- 1 executable scriptin 1 file: claudex-loop/scripts/runner.py:1
- 1 reference to secretsin 1 file: claudex-loop/scripts/runner.py:239
- No requests to vote in Hot
Show all 19 signals by file
.claude-plugin/marketplace.json
.claude-plugin/plugin.json
- Network accessline 4
"author": { "name": "Chase AI", "url": "https://youtube.com/@chaseai" - Network accessline 5
"name": "Chase AI", "url": "https://youtube.com/@chaseai" }, - Network accessline 6
"url": "https://youtube.com/@chaseai" - Network accessline 7
}, "homepage": "https://github.com/chaseai-yt/claudex-loop", "license": "MIT", - Network accessline 8
"homepage": "https://github.com/chaseai-yt/claudex-loop",
legacy/grill-me-codex/THIRD-PARTY-NOTICES.md
legacy/grill-with-docs-codex/THIRD-PARTY-NOTICES.md
- Network accessline 5
Act 1 of this skill ("GRILL WITH DOCS") and the companion files `CONTEXT-FORMAT.md` and `ADR-FORMAT.md` are adapted from the **grill-with-docs** skill by **Matt Pocock**, from https://github.com/mattpocock/skills, used under the MIT License - Network accessline 7
from https://github.com/mattpocock/skills, used under the MIT License.
skills/claudex-loop/references/runtime.md
- Network accessline 57
Primary references: [Claude programmatic usage](https://code.claude.com/docs/en/headless), [Claude CLI](https://code.claude.com/docs/en/cli-reference), [Codex non-interactive mode](https://learn.chatgpt.com/docs/non-interactive-mode). - Network accessline 57
Primary references: [Claude programmatic usage](https://code.claude.com/docs/en/headless), [Claude CLI](https://code.claude.com/docs/en/cli-reference), [Codex non-interactive mode](https://learn.chatgpt.com/docs/non-interactive-mode). - Network accessline 57
Primary references: [Claude programmatic usage](https://code.claude.com/docs/en/headless), [Claude CLI](https://code.claude.com/docs/en/cli-reference), [Codex non-interactive mode](https://learn.chatgpt.com/docs/non-interactive-mode).
skills/claudex-loop/scripts/runner.py
skills/claudex-route/SKILL.md
- Network accessline 40
For price-sensitive choices or comparative claims, consult current official [OpenAI model information](https://developers.openai.com/api/docs/models) and [Anthropic model information](https://platform.claude.com/docs/en/about-claude/models/ - Network accessline 54
Use the selected provider's CLI from the correct project directory, explicitly selecting the recommended or requested model for that call. Check the actual binary's version and help; consult the relevant official [Codex non-interactive guid
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: 19 facts
- Network access in code of the plugin configuration.claude-plugin/marketplace.json:3High confidence
- Network access in code of the plugin configuration.claude-plugin/marketplace.json:4High confidence
- Network access in code of the plugin configuration.claude-plugin/marketplace.json:5High confidence
- Network access in code of the plugin configuration.claude-plugin/plugin.json:4High confidence
- Network access in code of the plugin configuration.claude-plugin/plugin.json:5High confidence
- Network access in code of the plugin configuration.claude-plugin/plugin.json:6High confidence
- Network access in code of the plugin configuration.claude-plugin/plugin.json:7High confidence
- Network access in code of the plugin configuration.claude-plugin/plugin.json:8High confidence
- Network access in prose of documentationgrill-me-codex/THIRD-PARTY-NOTICES.md:5Low confidence
- Network access in prose of documentationgrill-me-codex/THIRD-PARTY-NOTICES.md:6Low confidence
- Network access in prose of documentationgrill-with-docs-codex/THIRD-PARTY-NOTICES.md:5Low confidence
- Network access in prose of documentationgrill-with-docs-codex/THIRD-PARTY-NOTICES.md:7Low confidence
- Network access in prose of documentationclaudex-loop/references/runtime.md:57Low confidence
- Network access in prose of documentationskills/claudex-loop/references/runtime.md:57Low confidence
- Network access in prose of documentationskills/claudex-loop/references/runtime.md:57Low confidence
- Executable script in code of a file that runsclaudex-loop/scripts/runner.py:1High confidence
- Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored valueclaudex-loop/scripts/runner.py:239Low confidence
- Network access in prose of the SKILL.mdclaudex-route/SKILL.md:40Low confidence
- Network access in prose of the SKILL.mdclaudex-route/SKILL.md:54Low confidence
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md legacy/grill-me-codex/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md legacy/grill-me-codex/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md legacy/grill-me-codex/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md legacy/grill-me-codex/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md legacy/grill-me-codex/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.
Explore alternatives and comparisons
Add this badge to your README
It updates every night and links to this page.
[](https://hotskills.dev/skill/chaseai-yt-claudex-loop/)