Skills for Real Engineers. Straight from my .agents directory.
retro
Agent skill fromhuytieu/COG-second-brain
CP-7 retrospective: audit checkpoints, evidence quality, action items, and harvest candidates. Closes the V-model cycle and feeds the next run. Use via /retro after ship, escalate, or significant session.
How to install retro
npx skills add huytieu/COG-second-brain# or add only this skill to Claude Code: copy the folder
skills/retro/
# into
~/.claude/skills/retro/ (all projects)
.claude/skills/retro/ (one project)
1# /retro — CP-7 close the V23Every non-`tiny` shipped or escalated run gets a retro. Retro is **advisory** but strongly expected: it is how the harness improves.45## When to run67- After CP-6 ship (approved publish, merged PR, vault deliverable done)8- After escalate (capture why verifier failed)9- End of an `/ultragoal` phase (one retro per phase)10- Manual: `/retro <run-dir or spec path>`1112## Phase 1 — Gather evidence bundle1314Read:15161. Spec or card description + acceptance criteria172. `04-projects/harness/runs/<id>/evidence/` (ledger, checkpoints.tsv, CP-* files)183. `.claude/logs/loop-ledger.tsv` (last rows for this run)194. `04-projects/<goal>/STATUS.md` if an ultragoal phase2021## Phase 2 — Checkpoint audit2223For each CP the lane required, mark Expected vs Actual vs Gap:2425| CP | `tiny` | `normal` | `full` |26|---|---|---|---|27| CP-1 spec | skip | required | required |28| CP-3v component | skip | required | required |29| CP-4 integration | skip | skip | required if multi-task |30| CP-5 acceptance | if mutate | required | required |31| CP-6 ship | if external | if external | required |3233## Phase 3 — Evidence quality pass3435Ask:3637- Any AC-n shipped without a PASS evidence row? (traceability violation)38- Any PASS row where observation is tool-return not artifact? (confident-but-unchecked)39- Any criterion that wasn't falsifiable? (fix in next spec)4041## Phase 4 — Write retro4243Copy `references/retro-template.md` to:4445`04-projects/harness/retro/YYYY-MM-DD-<slug>.md`46…
Excerpt shown under the repository's MIT license.
More skills in COG-second-brain
Similar skill repositories
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Production-grade engineering skills for AI coding agents.
The open-source app everyone uses to manage agents at work
The design vocabulary for your coding agent. Steer with polish and bolder, review with critique and audit, catch AI slop before it ships.
