retrospective-codify
Convert review feedback, CI failures, bug fixes, or task retrospectives into a durable project feedback ledger, then make future implementation work consult…
Skill by s-hiraoku
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/s-hiraoku-retrospective-codify
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: devops, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/retrospective-codify
What's inside
- Skill name
retrospective-codify- Description
- Convert review feedback, CI failures, bug fixes, or task retrospectives into a durable project feedback ledger, then make future implementation work consult that ledger before planning.
- License
- MIT
- Instructions
- 57 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
# Retrospective Codify Use this workflow when a task produced feedback that should change how future implementation starts. The goal is not to archive every comment; it is to extract reusable implementation rules from concrete review findings, CI failures, regressions, or user corrections. ## Workflow 1. Identify the feedback source. - Use PR review comments, CodeRabbit or Codex findings, CI failure analysis, bug reports, user corrections, or a completed task retrospective. - Separate one-off requests from reusable failure patterns. 2. Read the project feedback ledger. - Prefer `ledger/review-feedback.md` when present. - If the project uses another durable task-memory path, use that path and name it in the final report. - If no feedback ledger exists and the pattern is worth keeping, create a compact one. 3. Extract reusable lessons. - For each candidate, write the specific issue, root cause, general implementation rule, evidence, and verification signal. - Keep rules operational: future agents should know what to check before coding, not just what went wrong. - Do not store secrets, long pasted comments, private user data, or full logs. 4. Deduplicate before appending. - If an existing entry covers the same class of failure, update `Last seen`, `Evidence`, or `Implementation rule` instead of adding a parallel entry. - If the same pattern recurs, strengthen the rule so it is harder to miss during planning. 5. Feed the ledger into future work. - When starting implementation, read active feedback-ledger entries before planning. - Carry relevant active rules into the plan, tests, and self-review checklist. - If no entries apply, say that the ledger was checked and no relevant prior feedback was found. 6. Close or retire stale entries. - Mark an entry `Retired` only when project structure, tooling, or policy makes the rule obsolete. - Do not delete history unless it contains sensitive content that should not have been recorded. ## Entry Format Use this compact Markdown shape: ```md ### YYYY-MM-DD - Short Pattern Name - Status: Active - Last seen: YYYY-MM-DD - Source: PR review, CI failure, user correction, bug report, or retrospective - Trigger: when this kind of implementation work starts - Issue: what went wrong in the concrete task - Cause: why the original implementation missed it - Implementation rule: what future work must check or do before coding - Evidence: concise links, file paths, PR numbers, or command names - Verification: check that proves the rule was applied ``` ## Final Report Include: - feedback sources reviewed - ledger path updated or consulted - entries added, updated, skipped as one-off, or retired - implementation rules future work should apply - any sensitive details intentionally omitted
From s-hiraoku/codex-harnesses, MIT.
More from s-hiraoku/codex-harnesses
- 0.510score
pr-guardian
s-hiraoku
Monitor or resume GitHub pull requests, fix CI failures and actionable CodeRabbit, Codex, bot, or human review feedback, push focused updates, and stop only…
SkillBFreshness: BClaude Code +9 - 0.508score
autopilot
s-hiraoku
Run the PR Guardian workflow when the user explicitly invokes $autopilot to keep a GitHub pull request moving toward merge-ready without merging it.
SkillBFreshness: BClaude Code +9 - 0.465score
adviser
s-hiraoku
Route an independent review-only agent to a stronger model tier or reasoning effort at key decision points, then continue the task in the main agent.
SkillCFreshness: CClaude Code +9 - 0.465score
jina-read-url
s-hiraoku
Read, summarize, inspect, or extract content from a public URL through Jina Reader.
SkillCFreshness: CClaude Code +9 - 0.465score
no-slop
s-hiraoku
Strip AI slop from prose output and raise its information density — filler openings/closings, restated questions, contentless emphasis, stacked hedges…
SkillCFreshness: CClaude Code +9 - 0.462score
goal-manager
s-hiraoku
Manage /goal usage for concrete Codex work, including objective writing, success criteria, done criteria, long-running implementation, verification, PR…
SkillCFreshness: CClaude Code +9 - 0.460score
kaizen-loop
s-hiraoku
Run a continuous product improvement loop: evaluate a product or codebase end to end, identify product-quality improvement opportunities, present prioritized…
SkillCFreshness: CClaude Code +9 - 0.460score
implement-to-merge-ready
s-hiraoku
Run an end-to-end implementation delivery workflow from user request to merge-ready pull request.
SkillCFreshness: CClaude Code +9 - 0.460score
meta-packager
s-hiraoku
Analyze recent Codex work and package repeated patterns as the smallest useful skill, custom subagent, hook, or automation.
SkillCFreshness: CClaude Code +9 - 0.454score
deslop
s-hiraoku
Remove common AI-generated boilerplate and readability noise from changed code without changing behavior.
SkillCFreshness: CClaude Code +9 - 0.454score
fix-ci
s-hiraoku
Diagnose and repair failing CI checks on the current branch or pull request with focused, reproducible fixes.
SkillCFreshness: CClaude Code +9 - 0.433score
frontend-design
s-hiraoku
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 2
- Last commit (this folder)
- Jul 10, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/s-hiraoku/codex-harnesses
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from s-hiraoku
- 0.523score

Secondary Terminal
s-hiraoku
Sidebar terminal for VS Code with AI agent awareness (Claude Code, Copilot, Gemini, Codex, Antigravity), split views, session persistence, and full IME support.
PluginBFreshness: B★ 22VS Code +3 - 0.510score
pr-guardian
s-hiraoku
Monitor or resume GitHub pull requests, fix CI failures and actionable CodeRabbit, Codex, bot, or human review feedback, push focused updates, and stop only…
SkillBFreshness: BClaude Code +9 - 0.508score
autopilot
s-hiraoku
Run the PR Guardian workflow when the user explicitly invokes $autopilot to keep a GitHub pull request moving toward merge-ready without merging it.
SkillBFreshness: BClaude Code +9 - 0.505score
autopilot
s-hiraoku
Explicitly keep a GitHub pull request moving until it is merge-ready by delegating to PR Guardian.
SkillBFreshness: BClaude Code +9 - 0.465score
jina-reader
s-hiraoku
Fetch public web pages with Jina AI Reader API and convert them into LLM-friendly Markdown.
SkillCFreshness: CClaude Code +9 - 0.465score
adviser
s-hiraoku
Route an independent, review-only Claude process to a deterministically stronger model or effort when native Advisor is unavailable.
SkillCFreshness: CClaude Code +9
Cite this page
retrospective-codify skill. Agent Mart. https://agentmart.dev/skills/s-hiraoku-retrospective-codify (last updated 2026-10-24)