cc-hooks - SKILL.md Agent Skill

name: cc-hooks description: Run hook policy workflows.

cc-hooks (Codex)

Codex-native entry point for the cc-hooks operator skill.

The AgentOps source skill ../../skills/cc-hooks/SKILL.md is the source of truth for domain behavior, commands, examples, references, and output expectations. Read it first, then use prompt.md for the Codex runtime profile.

Codex Runtime Contract

Use Codex plus the local shell. Do not invoke Claude Code as an executor.
Load only the relevant source references or scripts for the task.
Prefer robot/JSON/NDJSON command surfaces when the source skill exposes them.
Verify command syntax from local --help or checked-in references before acting.
Return concrete evidence: commands run, files touched, exit codes, and any remaining blocker.

Opt-in guards

Installed-skill-edit guard — a PreToolUse Edit|Write guard that routes an edit of an installed skill copy (*/.claude/skills/**, .codex, .gemini) back to the repo source of truth skills/<name>/. Ships inert; activate with scripts/install-installed-skill-edit-guard.sh. See references/INSTALLED-SKILL-EDIT-GUARD.md.

Value-proof: on each fire the guard appends exactly one gate-blind JSONL line — {ts, session, token_class, path_sha256} — to ${AGENTOPS_HOME:-~/.agentops}/guardrail-telemetry.jsonl. The path is SHA-256 hashed, never raw (privacy); nothing is written on the happy path; the sensor is inert until installed. Pre-registered methodology (metric = declining fire-ATTEMPT rate over time; min N; null-at-small-N acceptable) satisfies ADR-0002 l.58. See references/GUARDRAIL-VALUE-PROOF.md.