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.
npx agentmods add skills/compozy/kb/impl-peer-reviewnpx skills add compozy/kb --skill impl-peer-reviewgit clone --depth 1 https://github.com/compozy/kbWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00213 | $0.04175 |
| Opus 5 | $0.00106 | $0.02087 |
| Sonnet 5 | $0.00043 | $0.00835 |
| Haiku 4.5 | $0.00021 | $0.00417 |
Grade A, and why
impl-peer-review 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.
How it starts
The opening of the file, as written. The whole thing — 332 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Peer Review
An independent model pressure-tests an implementation diff. This skill runs that cross-LLM pressure-test only when the user explicitly asks for a review round after an implementation pass. It is decoupled from any task-tracking system — the scope is the diff itself plus any optional context files the user names. The skill never auto-runs, auto-incorporates findings, auto-commits, or auto-loops additional rounds.
The skill is harness-agnostic and project-agnostic. It uses Compozy as the review engine:
each round runs one compozy exec against a configurable runtime (--ide/--model/
--reasoning). The review result is a direct-written Markdown findings file. compozy exec
stdout/stderr is operational evidence only; never parse it as the review source of truth.
Required Reading Router
Match your step to the row. Read the listed file in full before acting. These files are load-bearing, not appendices — the inline text in this SKILL.md is a pointer, not a substitute.
| Step | MUST read |
|---|---|
| Step 1.5 — gating readiness before review | references/readiness-checks.md |
| Step 2 — composing the review prompt | references/impl-review-prompt.md |
| Step 4 — validating findings | scripts/validate-findings.sh (run it) |
Reference Index
references/readiness-checks.md— the language-agnostic readiness markers that gate whether review credit is spent at all (build/test gate, non-empty diff, no stray artifacts, no WIP markers, generated-artifact co-ship, reviewable size). The highest-value forced read in this skill; the markers are not inlined here.references/impl-review-prompt.md— the canonical executable reviewer prompt template, including the scoped-write contract and the exact findings-file format. Its placeholders are substituted verbatim; never paraphrase it. (Over 100 lines — has its own## Contents.)scripts/validate-findings.sh— read-only structural validator for the findings artifact (frontmatter, required sections, no placeholders). Run it; never edit it.
What ships with it
3 files 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.
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.
- 2d ago First seen · 332 lines · 213 tokens per session scan A 098f863e7dde
impl-peer-review is a skill published in the GitHub repository compozy/kb (102 stars, last pushed 12d ago), licensed MIT. It adds 213 tokens to every session and 4,175 once invoked, about $0.0011 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
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…
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…
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…