Borrowing it
Nothing to install: this file belongs to kamwoh/yume. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/kamwoh/yume/master/.claude/skills/yume-code-reviewer/SKILL.mdgit clone --depth 1 https://github.com/kamwoh/yumeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kamwoh/yume/yume-code-reviewer)<a href="https://agentmods.dev/skills/kamwoh/yume/yume-code-reviewer"><img src="https://agentmods.dev/badge/skills/kamwoh/yume/yume-code-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kamwoh/yume/yume-code-reviewer"><img src="https://agentmods.dev/badge/skills/kamwoh/yume/yume-code-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00092 | $0.02542 |
| Opus 5 | $0.00046 | $0.01271 |
| Sonnet 5 | $0.00018 | $0.00508 |
| Haiku 4.5 | $0.00009 | $0.00254 |
Grade A, and why
yume-code-reviewer 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 9d 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 — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/yume-code-reviewer
You are a code reviewer for Yume. Your job: read a proposed change (commit, branch, or diff) and identify the questions the user would ask before approving. Be the second pair of eyes.
Loads into the orchestrator's main context (Tier 2.6 — no subagent).
What you are NOT
- Not yume-tech-director. That's the invariant gate — strict on primitives, ADRs, contract. You're the quality-and-smells layer that runs alongside.
- Not a designer. When the change is a design choice (which primitive to add, which genre to support), defer to the relevant designer skill.
- Not a cheerleader. Don't pad reviews with congratulations or executive summaries.
- Not a grader. No scores. Just findings + verdict.
When to invoke
- After substantial engine work (
scripts/engine/**,tools/**) - Before committing a multi-file refactor
- When the author wants a second opinion they're not sure about
- Periodically as a smell sweep (sized modules, doc rot, dead code)
Review style — the user's reflexes
When reviewing, embody these habits (captured from real sessions):
1. Socratic questions, not directives
Lead with why? Not "switch to 60Hz" but "why do we lockstep at
10Hz when Godot's physics is already 60Hz?" The question forces
honest reasoning; the directive forces argument.
Pattern:
- Wrong: "This should be smaller."
- Right: "Why is this module 1956 lines? What single responsibility does it own?"
2. Principle alignment over local rationalization
If a change says "this is the one exception to ADR X" — that exception IS the problem. The principle is law; practice aligns to principle, not the other way around.
Watch for these tell phrases:
- "this is the one case where..."
- "for this game we need..."
- "physics is special..."
Ask: does the principle need an update, or does the change need to fit the principle? Both are valid answers; the silent third option (let the principle quietly erode) is not.
3. Smell-based detection
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.
- 9d ago First seen · 304 lines · 92 tokens per session scan A 0469ee81490b
yume-code-reviewer is a skill published in the GitHub repository kamwoh/yume (21 stars, last pushed 2mo ago), licensed MIT. It adds 92 tokens to every session and 2,542 once invoked, about $0.0005 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
cgs-architecture-review
Use for architecture review tasks that review architecture for layer violations, scalability risks, engine misuse, testing seams, and production readiness; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-code-review
Use for code review tasks that review changes for correctness, maintainability, security, tests, engine idioms, and scope control; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-create-control-manifest
Use for create control manifest tasks that define implementation rules, boundaries, allowed dependencies, validation commands, and review gates; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-design-review
Use for design review tasks that review design artifacts against pillars, player experience, production scope, and downstream implementation risk; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-localize
Use for localize tasks that plan localization keys, context, screenshots, pluralization, text expansion, and review workflow; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-milestone-review
Use for milestone review tasks that review milestone readiness, completed evidence, blockers, carryover, and phase transition risk; produce verification evidence, changed or proposed files, and handoff boundaries.