[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78F5QF9XJFFMBAQSFWFFZ98":3,"skill-same:01M78F5QF9XJFFMBAQSFWFFZ98":144},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M78F5QF9XJFFMBAQSFWFFZ98","codex-review",null,"Runs OpenAI Codex as a local PR or uncommitted-diff reviewer and routes findings into the shared fix protocol. Triggers: '\u002Fcodex-review', 'codex review', 'run codex review', 'babysit codex review', 'codex as a PR reviewer'.","jl-cmd-projects\u002Fclaude-dev-env","jl-cmd-projects","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F338158757?v=4","01M78F4M55P4QY5YPN27XBQA9N","01M78F4M65AGEYC88NV6PW4715",7,"MIT","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: codex-review\ndescription: >-\n  Runs OpenAI Codex as a local PR or uncommitted-diff reviewer and routes\n  findings into the shared fix protocol. Triggers: '\u002Fcodex-review', 'codex\n  review', 'run codex review', 'babysit codex review', 'codex as a PR reviewer'.\n---\n\n# Codex Review\n\nRuns one Codex review pass against a chosen target, classifies the outcome, and hands findings to the shared fix protocol. Does not reimplement opt-out parsing or the fix sequence.\n\n## When this skill applies\n\n- The user wants Codex (the OpenAI Codex CLI) as a reviewer on the current PR branch, or on uncommitted work when no PR loop is active.\n- The user asks to babysit or re-run Codex review after fixes.\n\n## Refusals\n\nRespond with the quoted line exactly and stop:\n\n- Opt-out gate exit 0: `\u002Fcodex-review is disabled via CLAUDE_REVIEWS_DISABLED.`\n- Version or shape probe reports Codex unavailable: `\u002Fcodex-review cannot run: Codex CLI is missing or the shape probe failed.`\n- Skill class is `down` (wrapper `outcome_class` is `codex_down`): `\u002Fcodex-review cannot complete: Codex reviewer is down.`\n\nGate exits other than 0 or 1 are blockers: stop without a probe or review invoke. Do not invent an opt-out refusal for a non-opt-out failure.\n\n## Sub-skills\n\n| Skill | When | Produces |\n|---|---|---|\n| `reviewer-gates` | Step 0 — opt-out semantics for external reviewers | Gate contract for `reviews_disabled.py`; refusal line shape |\n| `pr-fix-protocol` | Step 5 — classification is `findings` | Fix sequence, reply-and-resolve unit, unresolved-thread sweep |\n\nIf `pr-fix-protocol` is not installed when findings exist, stop with: `\u002Fcodex-review needs the pr-fix-protocol skill to apply findings.`\n\n## Process checklist\n\n```\n- [ ] Step 0 — Opt-out gate\n- [ ] Step 1 — Version and shape probe\n- [ ] Step 2 — Target pick\n- [ ] Step 3 — Run classifying review\n- [ ] Step 4 — Classify outcome\n- [ ] Step 5 — Route findings (or stop on clean \u002F down)\n```\n\n### Step 0: Opt-out gate\n\nBefore any other work, run:\n\n```bash\npython \"$HOME\u002F.claude\u002F_shared\u002Fpr-loop\u002Fscripts\u002Freviews_disabled.py\" --reviewer codex\n```\n\n- **Exit 0** — Codex reviews are disabled: refuse with the opt-out line above. Do not probe, review, or fix.\n- **Exit 1** — continue.\n- **Any other exit** — treat as a blocker and stop; do not skip the gate or continue as if it exited 1. Do not report the opt-out refusal line for a non-opt-out failure.\n\nGate semantics live in `reviewer-gates` ([..\u002Freviewer-gates\u002FSKILL.md](..\u002Freviewer-gates\u002FSKILL.md)). The shared script owns the token parse; this skill does not re-parse `CLAUDE_REVIEWS_DISABLED`.\n\n### Step 1: Version and shape probe\n\nProbe the local Codex CLI so the run fails closed when the binary is missing or the installed shape does not match the contract.\n\n- **Contract owner:** [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md) — probe command (`codex exec review --help`), minimum shape signals, and fail-as-`codex_down` rule.\n- **Probe outcome:** continue only when the probe reports a supported shape. On failure, refuse with the probe refusal line above (skill class `down`).\n\n### Step 2: Target pick\n\nChoose what Codex reviews:\n\n| Context | Target |\n|---|---|\n| PR-loop caller (or an open PR on the current branch) | Diff against the PR base branch (`--base`) |\n| Standalone run with no PR | Uncommitted work via `--uncommitted` (staged + unstaged + untracked) |\n\nDo not invent a synthetic commit range when a base branch is available. Loop-caller wiring for base-branch resolution lives in [reference\u002Floop-integration.md](reference\u002Floop-integration.md).\n\n### Step 3: Run classifying review\n\nCall `scripts\u002Frun_codex_review.py` (`run_codex_review`) against the chosen target on the classifying path.\n\n- **Classifying command:** `codex exec [options] review --json …` — this is the only path that emits the success JSONL stream Step 4 reads. Full surface: [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md).\n- **Non-classifying form:** plain `codex review` (no `exec`, no `--json`) is not a classification input on the observed CLI; do not feed its stdout into Step 4.\n- **Contract owner:** [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md) — wrapper entrypoint, required arguments (target, repo root, run-state directory), capture fields, and exit codes.\n- **Wrapper boundary (capture only):** returns `completed` or `codex_down`, plus `exit_code`, `binary_version`, `jsonl_path`, and `agent_message`. It does not emit skill classes `down` \u002F `clean` \u002F `findings` and does not parse findings.\n- **This skill's job:** pass the Step 2 target, ensure `run_state_directory` exists, and hand the capture outcome to Step 4. Do not parse raw Codex stdout inline outside the documented parser path. Fail closed on probe miss, non-usable stream, or `codex_down`.\n\n### Step 4: Classify outcome\n\nMap the wrapper capture (and its JSONL observation) to exactly one skill-level class:\n\n| Skill class | From wrapper | Meaning | Next action |\n|---|---|---|---|\n| `down` | `outcome_class == codex_down` | Codex did not produce a usable review (tool failure, auth, crash, shape miss) | Refuse with the down line; stop |\n| `clean` | `outcome_class == completed` and findings parse empty | Review completed with no findings against the target | Report one-line clean summary; stop |\n| `findings` | `outcome_class == completed` and findings parse non-empty | Review completed with one or more addressable findings | Continue to Step 5 |\n\n`outcome_class` is the capture success signal. A process `exit_code` of 0 does not mean success when the wrapper already set `codex_down` (for example a shape probe that exits 0 but lacks required flags).\n\n#### Skill class ↔ CLI observation map\n\n| Skill class | Maps from (cli-contract) | Signal |\n|---|---|---|\n| `down` | `codex_down` rows; unrecognized probe shape | Non-usable review (exit 1\u002F2 failures, config\u002Fauth\u002Fmodel errors, probe miss) |\n| `clean` | Wrapper `completed` and `parse_codex_findings(agent_message)` empty (blank text or structured `[]`) | Usable review, zero addressable findings |\n| `findings` | Wrapper `completed` and `parse_codex_findings(agent_message)` non-empty (fenced JSON objects, freeform `- [P#]` bullets, or floor text) | Usable review with addressable findings |\n\nRaw CLI failure vocabulary is `codex_down` only ([reference\u002Fcli-contract.md](reference\u002Fcli-contract.md)). Skill steps and loop re-entry use `down` \u002F `clean` \u002F `findings`. Classification rules, findings parse, and payload fields live in [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md).\n\n### Step 5: Route findings into the shared fix protocol\n\nWhen the class is `findings`:\n\n1. Read `$HOME\u002F.claude\u002Fskills\u002Fpr-fix-protocol\u002FSKILL.md` (or invoke `pr-fix-protocol` by name when the `Skill` tool is available).\n2. Pass the findings payload, PR scope when present, and worktree path.\n3. Apply the protocol end to end — failing test first when behavior is at stake, one fix commit, push when the caller requires it, reply and resolve per thread when the review is on a PR.\n\nThis skill does not restate the fix sequence. Orchestrator callers that re-enter after a push follow [reference\u002Floop-integration.md](reference\u002Floop-integration.md).\n\n## Ground rules\n\n- **One capability:** run Codex review and classify; fixes go through `pr-fix-protocol`.\n- **Compose, do not rebuild:** opt-out via `reviews_disabled.py`; fixes via `pr-fix-protocol`.\n- **Fail closed** on opt-out, non-0\u002F1 gate exit, probe failure, and `down`.\n- **Preserve draft state** of any open PR; this skill does not flip ready.\n- **Honor hooks** on any commit the fix protocol creates.\n\n## Examples\n\n\u003Cexample>\nUser: `\u002Fcodex-review`\nClaude: [runs opt-out gate; on exit 1 probes Codex shape, picks base-branch or `--uncommitted` target, runs `codex exec … review --json`, classifies from JSONL, reports clean or routes findings; on non-0\u002F1 gate exit stops as a blocker]\n\u003C\u002Fexample>\n\n\u003Cexample>\n`CLAUDE_REVIEWS_DISABLED=codex`\nClaude: `\u002Fcodex-review is disabled via CLAUDE_REVIEWS_DISABLED.`\n\u003C\u002Fexample>\n\n\u003Cexample>\nUser: \"babysit codex review on this PR\"\nClaude: [same flow; after findings, applies pr-fix-protocol, then re-runs the classifying review once for confirmation when the caller stays on this skill]\n\u003C\u002Fexample>\n\n## Gotchas\n\n- **Opt-out uses `CLAUDE_REVIEWS_DISABLED=codex`.** `reviews_disabled.py --reviewer codex` exit 0 disables the review; exit 1 continues the probe and wrapper. Standalone `\u002Fcodex-review` stops with the opt-out refusal line and never runs the wrapper (no `codex_down` capture). Orchestrator gates map the same exit 0 to `codex_down` \u002F `codexDown` bypass on the machine checklist. Export `CLAUDE_REVIEWS_DISABLED=codex` (alone or comma-joined with other reviewers). A gate parse failure for an unknown token is a blocker — stop rather than skipping the gate. The shared gate must list `codex` among known reviewers for Step 0 to accept the flag.\n- **Only `codex exec … review --json` JSONL is classifiable.** Plain `codex review` stdout is non-classifying. Only a Step 4 `findings` class enters Step 5.\n- **Wrapper capture is not skill classification.** Step 5 consumes the skill-class payload from Step 4 (`down` \u002F `clean` \u002F `findings`), built from the wrapper's capture fields and the findings parse.\n- **Uncommitted targets have no review threads.** Fix protocol reply-and-resolve applies only when a PR carries threads; local-only runs stop after the fix commit path the protocol allows without a PR.\n\n## File index\n\n| File | Purpose |\n|---|---|\n| `SKILL.md` | Hub: opt-out, probe, target, classifying review, classify, fix handoff |\n| `CLAUDE.md` | Package map for agents opening this skill |\n| `reference\u002Fcli-contract.md` | CLI probe, wrapper capture, auth, cloud runbook, fixture-backed failure classes |\n| `reference\u002Floop-integration.md` | Gate placement, threshold, opt-out, state fields, re-entry |\n| `scripts\u002Frun_codex_review.py` | Headless capture wrapper (`run_codex_review`) |\n| `scripts\u002Ftest_run_codex_review.py` | Behavioral tests for the capture wrapper |\n| `scripts\u002Fcodex_down_classifier.py` | Maps exit code and stream text to detail class and `completed` \u002F `codex_down` |\n| `scripts\u002Fparse_codex_findings.py` | Parses agent text: fenced JSON, freeform bullets, then floor |\n| `scripts\u002Fcodex_review_scripts_constants\u002Frun_constants.py` | Importable named constants for the capture wrapper |\n| `scripts\u002Fcodex_usage_probe.py` | Weekly usage probe CLI (`percent_left`, `window_reset`, `source`) |\n| `scripts\u002Ftest_codex_usage_probe.py` | Probe unit tests (app-server fixtures and gate helper) |\n| `scripts\u002Fcodex_review_scripts_constants\u002Fcodex_usage_probe_constants.py` | Named constants for the weekly usage probe |\n\n## Folder map\n\n- `SKILL.md` — orchestration steps and refusals.\n- `CLAUDE.md` — purpose, trigger, key files.\n- `reference\u002F` — stable homes for cli-contract and loop-integration detail.\n- `scripts\u002F` — capture wrapper, classifier, findings parser, weekly usage probe, tests, fixtures, and constants package.\n- `scripts\u002Frun_codex_review.py` — headless capture wrapper entrypoint.\n- `scripts\u002Fcodex_down_classifier.py` — detail-class and gate-outcome classifier.\n- `scripts\u002Fparse_codex_findings.py` — findings parser entrypoint.\n- `scripts\u002Fcodex_usage_probe.py` — weekly usage probe entrypoint.\n- `scripts\u002Fcodex_review_scripts_constants\u002F` — importable constants package for skill scripts.\n","# Codex Review\n\nRuns one Codex review pass against a chosen target, classifies the outcome, and hands findings to the shared fix protocol. Does not reimplement opt-out parsing or the fix sequence.\n\n## When this skill applies\n\n- The user wants Codex (the OpenAI Codex CLI) as a reviewer on the current PR branch, or on uncommitted work when no PR loop is active.\n- The user asks to babysit or re-run Codex review after fixes.\n\n## Refusals\n\nRespond with the quoted line exactly and stop:\n\n- Opt-out gate exit 0: `\u002Fcodex-review is disabled via CLAUDE_REVIEWS_DISABLED.`\n- Version or shape probe reports Codex unavailable: `\u002Fcodex-review cannot run: Codex CLI is missing or the shape probe failed.`\n- Skill class is `down` (wrapper `outcome_class` is `codex_down`): `\u002Fcodex-review cannot complete: Codex reviewer is down.`\n\nGate exits other than 0 or 1 are blockers: stop without a probe or review invoke. Do not invent an opt-out refusal for a non-opt-out failure.\n\n## Sub-skills\n\n| Skill | When | Produces |\n|---|---|---|\n| `reviewer-gates` | Step 0 — opt-out semantics for external reviewers | Gate contract for `reviews_disabled.py`; refusal line shape |\n| `pr-fix-protocol` | Step 5 — classification is `findings` | Fix sequence, reply-and-resolve unit, unresolved-thread sweep |\n\nIf `pr-fix-protocol` is not installed when findings exist, stop with: `\u002Fcodex-review needs the pr-fix-protocol skill to apply findings.`\n\n## Process checklist\n\n```\n- [ ] Step 0 — Opt-out gate\n- [ ] Step 1 — Version and shape probe\n- [ ] Step 2 — Target pick\n- [ ] Step 3 — Run classifying review\n- [ ] Step 4 — Classify outcome\n- [ ] Step 5 — Route findings (or stop on clean \u002F down)\n```\n\n### Step 0: Opt-out gate\n\nBefore any other work, run:\n\n```bash\npython \"$HOME\u002F.claude\u002F_shared\u002Fpr-loop\u002Fscripts\u002Freviews_disabled.py\" --reviewer codex\n```\n\n- **Exit 0** — Codex reviews are disabled: refuse with the opt-out line above. Do not probe, review, or fix.\n- **Exit 1** — continue.\n- **Any other exit** — treat as a blocker and stop; do not skip the gate or continue as if it exited 1. Do not report the opt-out refusal line for a non-opt-out failure.\n\nGate semantics live in `reviewer-gates` ([..\u002Freviewer-gates\u002FSKILL.md](..\u002Freviewer-gates\u002FSKILL.md)). The shared script owns the token parse; this skill does not re-parse `CLAUDE_REVIEWS_DISABLED`.\n\n### Step 1: Version and shape probe\n\nProbe the local Codex CLI so the run fails closed when the binary is missing or the installed shape does not match the contract.\n\n- **Contract owner:** [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md) — probe command (`codex exec review --help`), minimum shape signals, and fail-as-`codex_down` rule.\n- **Probe outcome:** continue only when the probe reports a supported shape. On failure, refuse with the probe refusal line above (skill class `down`).\n\n### Step 2: Target pick\n\nChoose what Codex reviews:\n\n| Context | Target |\n|---|---|\n| PR-loop caller (or an open PR on the current branch) | Diff against the PR base branch (`--base`) |\n| Standalone run with no PR | Uncommitted work via `--uncommitted` (staged + unstaged + untracked) |\n\nDo not invent a synthetic commit range when a base branch is available. Loop-caller wiring for base-branch resolution lives in [reference\u002Floop-integration.md](reference\u002Floop-integration.md).\n\n### Step 3: Run classifying review\n\nCall `scripts\u002Frun_codex_review.py` (`run_codex_review`) against the chosen target on the classifying path.\n\n- **Classifying command:** `codex exec [options] review --json …` — this is the only path that emits the success JSONL stream Step 4 reads. Full surface: [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md).\n- **Non-classifying form:** plain `codex review` (no `exec`, no `--json`) is not a classification input on the observed CLI; do not feed its stdout into Step 4.\n- **Contract owner:** [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md) — wrapper entrypoint, required arguments (target, repo root, run-state directory), capture fields, and exit codes.\n- **Wrapper boundary (capture only):** returns `completed` or `codex_down`, plus `exit_code`, `binary_version`, `jsonl_path`, and `agent_message`. It does not emit skill classes `down` \u002F `clean` \u002F `findings` and does not parse findings.\n- **This skill's job:** pass the Step 2 target, ensure `run_state_directory` exists, and hand the capture outcome to Step 4. Do not parse raw Codex stdout inline outside the documented parser path. Fail closed on probe miss, non-usable stream, or `codex_down`.\n\n### Step 4: Classify outcome\n\nMap the wrapper capture (and its JSONL observation) to exactly one skill-level class:\n\n| Skill class | From wrapper | Meaning | Next action |\n|---|---|---|---|\n| `down` | `outcome_class == codex_down` | Codex did not produce a usable review (tool failure, auth, crash, shape miss) | Refuse with the down line; stop |\n| `clean` | `outcome_class == completed` and findings parse empty | Review completed with no findings against the target | Report one-line clean summary; stop |\n| `findings` | `outcome_class == completed` and findings parse non-empty | Review completed with one or more addressable findings | Continue to Step 5 |\n\n`outcome_class` is the capture success signal. A process `exit_code` of 0 does not mean success when the wrapper already set `codex_down` (for example a shape probe that exits 0 but lacks required flags).\n\n#### Skill class ↔ CLI observation map\n\n| Skill class | Maps from (cli-contract) | Signal |\n|---|---|---|\n| `down` | `codex_down` rows; unrecognized probe shape | Non-usable review (exit 1\u002F2 failures, config\u002Fauth\u002Fmodel errors, probe miss) |\n| `clean` | Wrapper `completed` and `parse_codex_findings(agent_message)` empty (blank text or structured `[]`) | Usable review, zero addressable findings |\n| `findings` | Wrapper `completed` and `parse_codex_findings(agent_message)` non-empty (fenced JSON objects, freeform `- [P#]` bullets, or floor text) | Usable review with addressable findings |\n\nRaw CLI failure vocabulary is `codex_down` only ([reference\u002Fcli-contract.md](reference\u002Fcli-contract.md)). Skill steps and loop re-entry use `down` \u002F `clean` \u002F `findings`. Classification rules, findings parse, and payload fields live in [reference\u002Fcli-contract.md](reference\u002Fcli-contract.md).\n\n### Step 5: Route findings into the shared fix protocol\n\nWhen the class is `findings`:\n\n1. Read `$HOME\u002F.claude\u002Fskills\u002Fpr-fix-protocol\u002FSKILL.md` (or invoke `pr-fix-protocol` by name when the `Skill` tool is available).\n2. Pass the findings payload, PR scope when present, and worktree path.\n3. Apply the protocol end to end — failing test first when behavior is at stake, one fix commit, push when the caller requires it, reply and resolve per thread when the review is on a PR.\n\nThis skill does not restate the fix sequence. Orchestrator callers that re-enter after a push follow [reference\u002Floop-integration.md](reference\u002Floop-integration.md).\n\n## Ground rules\n\n- **One capability:** run Codex review and classify; fixes go through `pr-fix-protocol`.\n- **Compose, do not rebuild:** opt-out via `reviews_disabled.py`; fixes via `pr-fix-protocol`.\n- **Fail closed** on opt-out, non-0\u002F1 gate exit, probe failure, and `down`.\n- **Preserve draft state** of any open PR; this skill does not flip ready.\n- **Honor hooks** on any commit the fix protocol creates.\n\n## Examples\n\n\u003Cexample>\nUser: `\u002Fcodex-review`\nClaude: [runs opt-out gate; on exit 1 probes Codex shape, picks base-branch or `--uncommitted` target, runs `codex exec … review --json`, classifies from JSONL, reports clean or routes findings; on non-0\u002F1 gate exit stops as a blocker]\n\u003C\u002Fexample>\n\n\u003Cexample>\n`CLAUDE_REVIEWS_DISABLED=codex`\nClaude: `\u002Fcodex-review is disabled via CLAUDE_REVIEWS_DISABLED.`\n\u003C\u002Fexample>\n\n\u003Cexample>\nUser: \"babysit codex review on this PR\"\nClaude: [same flow; after findings, applies pr-fix-protocol, then re-runs the classifying review once for confirmation when the caller stays on this skill]\n\u003C\u002Fexample>\n\n## Gotchas\n\n- **Opt-out uses `CLAUDE_REVIEWS_DISABLED=codex`.** `reviews_disabled.py --reviewer codex` exit 0 disables the review; exit 1 continues the probe and wrapper. Standalone `\u002Fcodex-review` stops with the opt-out refusal line and never runs the wrapper (no `codex_down` capture). Orchestrator gates map the same exit 0 to `codex_down` \u002F `codexDown` bypass on the machine checklist. Export `CLAUDE_REVIEWS_DISABLED=codex` (alone or comma-joined with other reviewers). A gate parse failure for an unknown token is a blocker — stop rather than skipping the gate. The shared gate must list `codex` among known reviewers for Step 0 to accept the flag.\n- **Only `codex exec … review --json` JSONL is classifiable.** Plain `codex review` stdout is non-classifying. Only a Step 4 `findings` class enters Step 5.\n- **Wrapper capture is not skill classification.** Step 5 consumes the skill-class payload from Step 4 (`down` \u002F `clean` \u002F `findings`), built from the wrapper's capture fields and the findings parse.\n- **Uncommitted targets have no review threads.** Fix protocol reply-and-resolve applies only when a PR carries threads; local-only runs stop after the fix commit path the protocol allows without a PR.\n\n## File index\n\n| File | Purpose |\n|---|---|\n| `SKILL.md` | Hub: opt-out, probe, target, classifying review, classify, fix handoff |\n| `CLAUDE.md` | Package map for agents opening this skill |\n| `reference\u002Fcli-contract.md` | CLI probe, wrapper capture, auth, cloud runbook, fixture-backed failure classes |\n| `reference\u002Floop-integration.md` | Gate placement, threshold, opt-out, state fields, re-entry |\n| `scripts\u002Frun_codex_review.py` | Headless capture wrapper (`run_codex_review`) |\n| `scripts\u002Ftest_run_codex_review.py` | Behavioral tests for the capture wrapper |\n| `scripts\u002Fcodex_down_classifier.py` | Maps exit code and stream text to detail class and `completed` \u002F `codex_down` |\n| `scripts\u002Fparse_codex_findings.py` | Parses agent text: fenced JSON, freeform bullets, then floor |\n| `scripts\u002Fcodex_review_scripts_constants\u002Frun_constants.py` | Importable named constants for the capture wrapper |\n| `scripts\u002Fcodex_usage_probe.py` | Weekly usage probe CLI (`percent_left`, `window_reset`, `source`) |\n| `scripts\u002Ftest_codex_usage_probe.py` | Probe unit tests (app-server fixtures and gate helper) |\n| `scripts\u002Fcodex_review_scripts_constants\u002Fcodex_usage_probe_constants.py` | Named constants for the weekly usage probe |\n\n## Folder map\n\n- `SKILL.md` — orchestration steps and refusals.\n- `CLAUDE.md` — purpose, trigger, key files.\n- `reference\u002F` — stable homes for cli-contract and loop-integration detail.\n- `scripts\u002F` — capture wrapper, classifier, findings parser, weekly usage probe, tests, fixtures, and constants package.\n- `scripts\u002Frun_codex_review.py` — headless capture wrapper entrypoint.\n- `scripts\u002Fcodex_down_classifier.py` — detail-class and gate-outcome classifier.\n- `scripts\u002Fparse_codex_findings.py` — findings parser entrypoint.\n- `scripts\u002Fcodex_usage_probe.py` — weekly usage probe entrypoint.\n- `scripts\u002Fcodex_review_scripts_constants\u002F` — importable constants package for skill scripts.\n",{"name":5,"description":7},"main","974284b42eeb4609e94d4ebc1f6e36df400d54ca","packages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review",1,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects","2026-11-15 05:02:47",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add jl-cmd-projects\u002Fclaude-dev-env --skill codex-review",{"tool":36,"command":37},"bunx","bunx skills add jl-cmd-projects\u002Fclaude-dev-env --skill codex-review",{"tool":39,"command":40},"pnpm","pnpm dlx skills add jl-cmd-projects\u002Fclaude-dev-env --skill codex-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002FSKILL.md",true,[44,50,54,60,64,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"reference\u002Fcli-contract.md","REFERENCE","参考文档",13925,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Freference\u002Fcli-contract.md",{"path":51,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":52,"sourceHtmlUrl":53},"reference\u002Floop-integration.md",7062,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Freference\u002Floop-integration.md",{"path":55,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":58,"sourceHtmlUrl":59},"scripts\u002Fcodex_down_classifier.py","SCRIPT","脚本",3242,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_down_classifier.py",{"path":61,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":62,"sourceHtmlUrl":63},"scripts\u002Fcodex_review_scripts_constants\u002F__init__.py",58,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_review_scripts_constants\u002F__init__.py",{"path":65,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":66,"sourceHtmlUrl":67},"scripts\u002Fcodex_review_scripts_constants\u002Fclassifier_constants.py",848,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_review_scripts_constants\u002Fclassifier_constants.py",{"path":69,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":70,"sourceHtmlUrl":71},"scripts\u002Fcodex_review_scripts_constants\u002Fcodex_usage_probe_constants.py",2412,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_review_scripts_constants\u002Fcodex_usage_probe_constants.py",{"path":73,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":74,"sourceHtmlUrl":75},"scripts\u002Fcodex_review_scripts_constants\u002Ffindings_constants.py",530,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_review_scripts_constants\u002Ffindings_constants.py",{"path":77,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":78,"sourceHtmlUrl":79},"scripts\u002Fcodex_review_scripts_constants\u002Frun_constants.py",1748,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_review_scripts_constants\u002Frun_constants.py",{"path":81,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":82,"sourceHtmlUrl":83},"scripts\u002Fcodex_usage_probe.py",19375,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fcodex_usage_probe.py",{"path":85,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":86,"sourceHtmlUrl":87},"scripts\u002Ffixtures\u002Fauth_failure_synthetic.txt",71,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fauth_failure_synthetic.txt",{"path":89,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":90,"sourceHtmlUrl":91},"scripts\u002Ffixtures\u002Fconfig_load_failure_v0.125.0.txt",98,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fconfig_load_failure_v0.125.0.txt",{"path":93,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":94,"sourceHtmlUrl":95},"scripts\u002Ffixtures\u002Ffreeform_findings_v0.144.3.txt",420,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Ffreeform_findings_v0.144.3.txt",{"path":97,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":98,"sourceHtmlUrl":99},"scripts\u002Ffixtures\u002Fmodel_rejection_v0.125.0.jsonl",573,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fmodel_rejection_v0.125.0.jsonl",{"path":101,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":102,"sourceHtmlUrl":103},"scripts\u002Ffixtures\u002Fstructured_findings.txt",250,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fstructured_findings.txt",{"path":105,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":106,"sourceHtmlUrl":107},"scripts\u002Ffixtures\u002Fsuccess_stream_v0.144.3.jsonl",585,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fsuccess_stream_v0.144.3.jsonl",{"path":109,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":110,"sourceHtmlUrl":111},"scripts\u002Ffixtures\u002Funknown_failure_synthetic.txt",57,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Funknown_failure_synthetic.txt",{"path":113,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":114,"sourceHtmlUrl":115},"scripts\u002Ffixtures\u002Fusage_limit_synthetic.txt",95,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ffixtures\u002Fusage_limit_synthetic.txt",{"path":117,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":118,"sourceHtmlUrl":119},"scripts\u002Fparse_codex_findings.py",6773,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Fparse_codex_findings.py",{"path":121,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":122,"sourceHtmlUrl":123},"scripts\u002Frun_codex_review.py",21383,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Frun_codex_review.py",{"path":125,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":126,"sourceHtmlUrl":127},"scripts\u002Ftest_codex_down_classifier.py",4637,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ftest_codex_down_classifier.py",{"path":129,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":130,"sourceHtmlUrl":131},"scripts\u002Ftest_codex_usage_probe.py",23245,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ftest_codex_usage_probe.py",{"path":133,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":134,"sourceHtmlUrl":135},"scripts\u002Ftest_parse_codex_findings.py",4004,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ftest_parse_codex_findings.py",{"path":137,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":138,"sourceHtmlUrl":139},"scripts\u002Ftest_run_codex_review.py",32155,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Fscripts\u002Ftest_run_codex_review.py",{"path":141,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":142,"sourceHtmlUrl":143},"test_skill_scaffold.py",7235,"https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fpackages\u002Fclaude-dev-env\u002F.agents\u002Fskills-archived\u002Fcodex-review\u002Ftest_skill_scaffold.py",[145,152,159,166,173,180],{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":18},"01M78FER7S09AE87MMJFT1E87E","update","Fast-forwards a local repository's main branch to a remote's main after confirming the repo path and source remote through AskUserQuestion. Use when the user says '\u002Fupdate', 'update main', 'fast-forward main', 'sync main from origin', 'pull latest main into \u003Cpath>', or 'bring main up to date'.","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Fupdate\u002FSKILL.md",[],[],{"skillNo":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":158,"publishedAt":18},"01M78FENWZS01MB4PMGX6801X6","source-command-sr-loop","Converging cleanup loop: repeat \u002Fsimplify until clean, then repeat \u002Fcode-review --fix until clean","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Fsource-command-sr-loop\u002FSKILL.md",[],[],{"skillNo":160,"name":161,"nameZh":6,"description":162,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":163,"categories":164,"keywords":165,"publishedAt":18},"01M78FEGZ7JDWEG3RM78M9Y8F8","skill-builder","Skill lifecycle: classify, scaffold, write via the skill-writer-agent, self-audit, compose sub-skills, polish description triggers, enforce deterministic scripts. Triggers: build a skill, new skill workflow, improve this skill, optimize skill description, skill development lifecycle, skill modularity, skill composition, description trigger catalog, deterministic skill scripts.","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Fskill-builder\u002FSKILL.md",[],[],{"skillNo":167,"name":168,"nameZh":6,"description":169,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":170,"categories":171,"keywords":172,"publishedAt":18},"01M78FEDEYHJYKTER9S566DMEZ","session-tidy","Audit, clean, and consolidate session logs in the Obsidian vault. Fixes format drift, resolves orphaned next-steps, updates stale statuses, and generates project rollup summaries. Use when session logs feel cluttered or before starting a new project milestone. Triggers on '\u002Fsession-tidy', 'tidy sessions', 'clean up session logs', 'session audit'.","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Fsession-tidy\u002FSKILL.md",[],[],{"skillNo":174,"name":175,"nameZh":6,"description":176,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":177,"categories":178,"keywords":179,"publishedAt":18},"01M78FE8QR789SKBF3A78N2T13","session-log","Logs a session report as an HTML page, tracks vault context, extracts decisions, and outputs a \u002Frename command. Use for \u002Fsession-log, 'journal this session', 'log this work', 'session report', 'summarize\u002Flog\u002Frecord this session', 'save session', 'capture session', or 'document what we did'.","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Fsession-log\u002FSKILL.md",[],[],{"skillNo":181,"name":182,"nameZh":6,"description":183,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":184,"categories":185,"keywords":186,"publishedAt":18},"01M78FE4JDJHKTJPWRWNQEX2HT","run-claude-dev-env","Build, install, run, and test the claude-dev-env installer. Use when asked to run claude-dev-env, install the Claude Code config into ~\u002F.claude, drive the installer, smoke-test it against a sandbox, or run its test suite.","https:\u002F\u002Fgithub.com\u002Fjl-cmd-projects\u002Fclaude-dev-env\u002Fblob\u002Fmain\u002Fskill-archive\u002Frun-claude-dev-env\u002FSKILL.md",[],[]]