retro

A measured review of the development process that studies past incidents, rework, usage records, and tested improvement rules.

In plain words
What is it for?
It is for auditing project records, grouping errors, measuring costs and delays, reviewing benchmark cases, and proposing process changes for human approval.
Why use it?
It turns recurring problems into repeatable checks and avoids adopting process changes that have not been shown to help.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/wallkop/opc-develop/retro
Any agent
npx skills add wallkop/opc-develop --skill retro
Clone the repo
git clone --depth 1 https://github.com/wallkop/opc-develop

Made for: Claude Code, Codex.

Per session 40 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 658 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00040 $0.00658
Opus 5 $0.00020 $0.00329
Sonnet 5 $0.00008 $0.00132
Haiku 4.5 $0.00004 $0.00066

Measured 2d ago against content hash b7c0cd3d740e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

retro scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/retro/SKILL.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

retro

The loop that improves the loop. A historical problem becomes a cheap repeatable experiment; a rule becomes active only after that experiment proves it can kill the bad variant.

Load

  • ${CLAUDE_PLUGIN_ROOT}/shared/core-contract.md
  • ${CLAUDE_PLUGIN_ROOT}/shared/formats/ledger-format.md
  • ${CLAUDE_PLUGIN_ROOT}/shared/formats/benchmark-case-format.md
  • ${CLAUDE_PLUGIN_ROOT}/shared/formats/report-style.md

Process

  1. Data-quality preflight: audit every active feature/error ledger and generated acceptance receipt. Count eligible phase/gate/dispatch entries, cost coverage, unmapped sessions, invalid gaps, command/provider capture, and benchmark linkage. Missing cost data creates or updates one stable observe gap with its label cap; never duplicate it.
  2. Collect: feature summaries, exact-match error pre-groups, semantic clusters, normalized Codex/Claude usage, benchmark reports, and crystallized rules. Label script findings separately from semantic judgment.
  3. Compute: measured/no-data cost by phase, time to first runnable core path, command and repeated-test counts, subtask/context modes, review/repair rounds, provider calls/cost, rework routing, user corrections, post-completion fixes, recurring clusters, open gaps, benchmark coverage, and false-green gates. Do not call missing capture “zero recurrence” or “efficient.”
  4. Executable case check: every P0/P1, irreversible-risk, false-green, or recurring incident links a project case, or a human-approved waiver. Profiles reproduce one problem at different cost. Prefer the cheapest stable profile and report when real calibration is stale.
  5. Crystallize: propose the lowest enforcement layer. Approval records approved; building records implemented; only a linked GREEN → RED → GREEN report promotes verified, then active-unmeasured until later observations exist.
  6. Measure: after approval, count linked recurrences and rule fires. Zero with healthy capture becomes effective; recurrence becomes recurring and proposes a lower layer; 8 weeks unfired or a model major-version change becomes a retirement candidate.
  7. Human report: write docs/opc/retro/<date>.json and .md, then render/lint docs/opc/retro/<date>.html. Lead with conclusion, user impact, evidence, and decision. Only human-approved rule changes are written to docs/opc/rules.md.

Read the full file on GitHub · 55 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 55 lines · 40 tokens per session scan A b7c0cd3d740e

Subscribe to this mod's changes

retro is a skill published in the GitHub repository wallkop/opc-develop (11 stars, last pushed 1mo ago), licensed MIT. It adds 40 tokens to every session and 658 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens