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 skills add alphacornutum/layercake --skill commentsgit clone --depth 1 https://github.com/alphacornutum/layercakeWrote 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/alphacornutum/layercake/comments)<a href="https://agentmods.dev/skills/alphacornutum/layercake/comments"><img src="https://agentmods.dev/badge/skills/alphacornutum/layercake/comments/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/alphacornutum/layercake/comments"><img src="https://agentmods.dev/badge/skills/alphacornutum/layercake/comments.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 | $0.00089 | $0.00640 |
| Opus 5.5 | $0.00036 | $0.00256 |
| Sonnet 5.5 | $0.00018 | $0.00128 |
| Haiku 4.5 | $0.00009 | $0.00064 |
Grade A, and why
comments 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 20d 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Comments Cleanup
Scan the diff you just produced and strip comments that the code already carries. The comments rule has the long-form criteria and examples; this skill is the cleanup procedure.
Procedure
- Pull the diff.
git difffor unstaged work,git diff --stagedfor staged. Focus on+-lines that introduced comments. - For each new comment, ask one question — would a competent reader miss anything if this comment were removed? If the answer is no, delete the comment.
- Tighten what remains. A comment that stays should describe why, fit on one line where possible, and use the language's doc-comment syntax for public APIs only.
Categories to delete on sight
- Narration — restates what the code does (
// increment counter,// loop through users,// validate user). - Step markers — planning artefacts (
// Step 1: fetch users,// now we will…,// AI thought: …). - Task / PR / caller references (
// added for ticket X,// used by Y) — that context belongs in the commit message. - Commented-out code — git keeps the history.
- Decorative banners (
// ===== HELPERS =====) when nothing else in the file uses them. - Apologies (
// hacky but works) — either fix the code or open a tracked issue and link it. - Untracked
TODO/FIXME— pair every marker with an owner or an issue, or resolve it now.
Categories to keep
- Hidden constraint — a non-obvious invariant, ordering requirement, or performance assumption.
- External quirk — API behaviour, browser bug, platform-specific oddity (name the system).
- Workaround — a fix for a tracked upstream issue (link it).
- Surprise — behaviour that would make a reasonable reader pause.
- Public API contract — inputs, outputs, errors, side effects via
///,/** */, or docstrings.
Self-check before finishing
- Diff scanned end-to-end, including test files.
- No
// Step N, no// loop, no// AI thought, no commented-out blocks. - Every remaining
TODO/FIXMEhas an owner or issue. - Comment density matches the surrounding files (scan 5–10 nearby).
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.
- 20d ago First seen · 42 lines · 89 tokens per session scan A abde92d97857
comments is a skill published in the GitHub repository alphacornutum/layercake (0 stars, last pushed 2d ago), licensed MIT. It adds 89 tokens to every session and 640 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-26.
Other skills, from other repositories
after-effects-scripting
Drive a running After Effects instance directly from the shell to build, edit and render motion graphics — no MCP, plugin or panel required. Use whenever the user wants Claude to work inside After Effects: building animations, title sequences, logo stings, UI/product motion, data visualisations, or fixing an existing…
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…