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/progiri/craftlight/code-reviewnpx skills add progiri/craftlight --skill code-reviewgit clone --depth 1 https://github.com/progiri/craftlightWhat 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 | $0.00253 | $0.01731 |
| Opus 5 | $0.00127 | $0.00865 |
| Sonnet 5 | $0.00051 | $0.00346 |
| Haiku 4.5 | $0.00025 | $0.00173 |
Grade A, and why
code-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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
code-review — review without edits
Three principles on top of the shared ones: (1) a review reads and proves but doesn't edit; (2) every finding is a file:line proof plus a failure scenario; (3) a false positive costs more than a missed nit — it burns trust in the whole report, so the unproven doesn't go into the report.
Not for this skill: "do/fix/add" — that's task; a known breakage whose cause is unknown → debug; "how does this work" and onboarding docs — codebase-reverse-engineering. Fixing what's found is task again: the review hands it the findings as a checklist.
Step 0. The review object
Determine the object with cheap commands (git status/diff/log, gh pr view/diff); an object explicitly named by the user beats the defaults:
- the uncommitted diff — the default, if there is one: staged + unstaged + untracked non-ignored files (
git status --short, not a baregit diff— a new file is often the riskiest part of a change, andgit diffis blind to it); - a branch against its base (merge-base with main/master);
- a PR —
gh pr diff; - a range of commits;
- a module/files or a snippet from the chat — an audit of code without a diff.
Exclude generated and vendored files (lock files, dist/build, snapshots, minified) right away: they aren't reviewed line by line, they're mentioned in one line. Then announce the object and size with the junk already removed: Object: branch task/search against main — 7 files, ~240 lines. No diff and no object named → one concrete question.
The craftlight block upsert (reference and procedure — skills/task/templates/CLAUDE-block.md) rides with the only file this skill ever writes — the Full-mode report. Express writes nothing → it must not edit the user's CLAUDE.md; a review is atomic, and an interrupted one is re-run rather than resumed (no resume machinery — the deliberate exception in the family). The one exception: a compact-tripwire draft report (Status: draft, full.md) — after the reset, finish its missing passes instead of re-running the done ones.
What ships with it
4 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 · 62 lines · 253 tokens per session scan A ff41bdfad7fa
code-review is a skill published in the GitHub repository progiri/craftlight (5 stars, last pushed 8d ago), licensed MIT. It adds 253 tokens to every session and 1,731 once invoked, about $0.0013 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 2026-08-31.
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.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…