菜单
Run ADLC P7 distillation and maintenance workflows in Pi, including lesson-foundry, rejection-mining, skill-mining, scheduled maintenance, skill-rot, model-ratchet, review-calibration, and gate-fuzzing.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Refute-charter review discipline for prosecuting a diff against its ticket. Use when asked to act as a prosecutor, reviewer, or to find reasons a change must not merge.
Run ADLC P7 distillation and maintenance workflows in Codex, including lesson-foundry, rejection-mining, skill-mining, scheduled maintenance, skill-rot, model-ratchet, review-calibration, and gate-fuzzing.
Run the multi-lens P5 adversarial prosecution fan-out and record ADLC P5 review evidence and P6 acceptance packet workflows in Codex. Fans out five independent read-only lenses, dedupes findings, verifies each independently, and loops until dry before recording evidence.
Run ADLC P7 distillation in the GitHub Copilot CLI — lesson-foundry, rejection-mining, and review-calibration — turning repeated review findings and PR rejections into permanent, deterministic defenses. Decay-driven maintenance lives in the adlc-maintain skill.
Run the decay-driven ADLC maintenance checks — stale skills, hot files to re-prosecute, stale tickets, and gate calibration (C10/C12).
Run the multi-lens P5 adversarial prosecution fan-out and record ADLC P5 review evidence and P6 acceptance packet workflows in the GitHub Copilot CLI. Fans out five independent read-only lenses, dedupes findings, verifies each independently, and loops until dry before recording evidence.
| name | adlc-distill |
| description | Run ADLC P7 distillation and maintenance workflows in Pi, including lesson-foundry, rejection-mining, skill-mining, scheduled maintenance, skill-rot, model-ratchet, review-calibration, and gate-fuzzing. |
P7 converts repeated findings into deterministic defenses and keeps cached guidance fresh.
Commands:
adlc lesson-foundry --json
adlc rejection-mining --json
adlc skill-rot .agents/skills plugins/adlc-pi/skills --json
adlc model-ratchet --json
adlc review-calibration --review-cmd "npx adversarial-review --base {base}" --json
adlc gate-fuzzing --json
Scheduled or automated P7 maintenance should invoke this skill and the external $skill-mining workflow when available, with CI cron as the deterministic fallback.
Record a no-op manifest entry when there is nothing to distill so the runner can distinguish "checked and empty" from "skipped."
Whether you just distilled findings or ran the maintenance commands above, finish on
the ledger's own gate — it must exit 0:
adlc lesson-foundry --gate
A non-zero exit names each cluster in .adlc/findings.jsonl that recurs but still
has no banked lesson; each is an undefended cluster. Bank or refine its lesson and
re-run until the gate is green. It is deterministic, keyless, and portable:
.adlc/findings.jsonl is tracked in git (ADR 0014), so the cluster set travels with
the repo and the gate returns the same verdict anywhere it is checked out. It is not
yet wired into the maintenance cron.