Free SKILL.md scraped from GitHub. Clone the repo or copy the file directly into your Claude Code skills directory.
npx versuz@latest install joshua-eisenhart-codex-ratchet-system-v4-skill-specs-a2-skill-source-intake-operatorgit clone https://github.com/Joshua-Eisenhart/Codex-Ratchet.gitcp Codex-Ratchet/SKILL.MD ~/.claude/skills/joshua-eisenhart-codex-ratchet-system-v4-skill-specs-a2-skill-source-intake-operator/SKILL.md--- skill_id: a2-skill-source-intake-operator name: a2-skill-source-intake-operator description: Run the first bounded A2-side source-intake audit for the broad skill-source corpus, with special focus on the first imported `lev-os/agents` intake/discovery/build cluster and whether the front-door corpus, A2 indexing, and local imported-skill reality still agree. skill_type: maintenance source_type: repo_skill applicable_layers: [A2_HIGH_INTAKE, A2_MID_REFINEMENT] applicable_graphs: [runtime, control, a2_high_intake_graph_v1] inputs: [repo_root, report_path] outputs: [intake_report] related_skills: [autoresearch-operator, skill-improver-operator, witness-evermem-sync] capabilities: can_write_repo: true can_only_propose: false tool_dependencies: [] provenance: "lev-os/agents lev-intake + skill-discovery + skill-builder, retooled into a bounded A2 intake audit slice" adapters: codex: system_v4/skill_specs/a2-skill-source-intake-operator/SKILL.md shell: system_v4/skills/a2_skill_source_intake_operator.py --- # A2 Skill Source Intake Operator Use this skill when the broad skill-source corpus, imported skill clusters, and canonical A2 indexing need a bounded truth refresh. ## Purpose - keep the broad source corpus from drifting away from local repo reality - keep imported-cluster promotion grounded in actual downloaded sources - keep canonical A2 indexing honest for the front-door corpus docs - produce one structured intake report instead of more vague tracker prose ## Execute Now 1. Read: - [SKILL_SOURCE_CORPUS.md](/home/ratchet/Desktop/Codex%20Ratchet/SKILL_SOURCE_CORPUS.md) - [REPO_SKILL_INTEGRATION_TRACKER.md](/home/ratchet/Desktop/Codex%20Ratchet/REPO_SKILL_INTEGRATION_TRACKER.md) - [SKILL_CANDIDATES_BACKLOG.md](/home/ratchet/Desktop/Codex%20Ratchet/SKILL_CANDIDATES_BACKLOG.md) - [LOCAL_SOURCE_REPO_INVENTORY.md](/home/ratchet/Desktop/Codex%20Ratchet/LOCAL_SOURCE_REPO_INVENTORY.md) - [A2_SKILL_SOURCE_INTAKE_PROCEDURE__CURRENT__v1.md](/home/ratchet/Desktop/Codex%20Ratchet/system_v3/a2_state/A2_SKILL_SOURCE_INTAKE_PROCEDURE__CURRENT__v1.md) 2. Audit the first imported cluster: - `lev-intake` - `skill-discovery` - `skill-builder` 3. Verify: - front-door corpus doc presence - canonical A2 direct indexing for root corpus docs - local `lev-os/agents` skill counts - first imported-cluster map presence 4. Emit one structured report. 5. Default repo-held outputs when no explicit path is supplied: - `system_v4/a2_state/audit_logs/A2_SKILL_SOURCE_INTAKE_REPORT__CURRENT__v1.json` - `system_v4/a2_state/audit_logs/A2_SKILL_SOURCE_INTAKE_REPORT__CURRENT__v1.md` ## Quality Gates - Do not claim imported-cluster live integration from source presence alone. - Do not import workshop/toolchain assumptions like `.lev/workshop`, `qmd`, or `skill-seekers`. - Keep the first slice bounded to inventory, classification, and staged proposal output. - If a count or path mismatches, record it explicitly instead of smoothing it away.