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 instructions/davidcjw/skill-rank/agents-mdgit clone --depth 1 https://github.com/davidcjw/skill-rankWrote 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/instructions/davidcjw/skill-rank/agents-md)<a href="https://agentmods.dev/instructions/davidcjw/skill-rank/agents-md"><img src="https://agentmods.dev/badge/instructions/davidcjw/skill-rank/agents-md.svg" alt="Measured on agentmods" 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 | $0.00910 | $0.00910 |
| Opus 5 | $0.00455 | $0.00455 |
| Sonnet 5 | $0.00182 | $0.00182 |
| Haiku 4.5 | $0.00091 | $0.00091 |
Grade A, and why
skill-rank AGENTS.md 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 4d 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.
AGENTS.md — SkillRank
Guidance for AI agents working in this repo.
What this is
A Next.js 16 app that ranks Claude / AI-agent skills by a cross-web popularity score and re-ranks every 24 hours. No database required to run; an optional Supabase table unlocks day-over-day movement arrows.
Commands
npm run dev # local dev server
npm test # vitest (ranking engine) — run before claiming a feature done
npm run lint # eslint (next core-web-vitals + ts)
npm run build # production build; must pass before deploy
Always run npm test and npm run lint before considering a change complete.
Where things live
- Scoring logic is in
lib/rank.tsand is pure (no I/O). All ranking/merge changes go here and must keeplib/__tests__/rank.test.tsgreen. Add a test for any new signal or weight. - Each web source is one file in
lib/sources/exporting a singlefetch*Candidates()that must be resilient — wrap everything in try/catch and return[]on failure. Never let a dead source break the page. Current sources:github,npm,hackernews,awesomeLists(extractsgh:owner/repolinks from curated "awesome-*" READMEs via raw.githubusercontent.com →mentionsendorsement signal, community bucket). A "buzz/mention" source only adds value if it maps to a canonicalgh:owner/repokey, since merge is by key. lib/trending.ts(selectTrending/hasTrending) is a pure selector of risers/fallers/new from the ranked board'sdelta/isNew.components/Trending.tsxrenders the strip and returnsnulluntil movement data exists. Keep the selector tested.- Source fetches must be tagged for cache invalidation:
next: { revalidate: 86_400, tags: ["skills"] }. The cron route callsrevalidateTag("skills"). lib/skills.tsis the orchestrator. New sources are wired in here and intoliveSources. It filters live candidates throughisRelevantSkillbefore ranking.lib/relevance.tsis the on-topic gate: an allowlist of Claude/agent terms (RELEVANCE_PATTERNS) plus a curated denylist of non-skill repos (EXCLUDE_KEY_PATTERNS, e.g. n8n, vendor CLIs). When a general-purpose tool keeps surfacing, add it to the denylist and cover it inlib/__tests__/relevance.test.ts.lib/seed.tsis the offline fallback; keep entries real and plausible.- UI lives in
components/. The board is OLED-dark, Fira Code/Sans, with tabular figures (.tnum) on every number. Keep that aesthetic.
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.
- 4d ago First seen · 42 lines · 910 tokens per session scan A ddd8565b3535
skill-rank AGENTS.md is an instructions file published in the GitHub repository davidcjw/skill-rank (2 stars, last pushed 2mo ago), licensed MIT. It adds 910 tokens to every session, about $0.0046 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 instructions, from other repositories
sigrank-app AGENTS.md
AGENTS.md instructions for SunrisesIllNeverSee/sigrank-app, covering agents.md — sigrank-app, quick reference, verification protocol (before every commit), frozen invariants (never change) and seo/aeo/geo content (do not touch).
core CLAUDE.md
Claude Code instructions for module-federation/core, covering claude task parallelization & development guidelines, core principles, maximum parallelization strategy, universal parallel task patterns and for any codebase task.
mastra AGENTS.md
AGENTS.md instructions for mastra-ai/mastra: Unless asked, don't inspect reference or modify examples. Use the most-specific AGENTS.md; for package work, read packages/ /AGENTS.md first.
souls-directory CLAUDE.md
Instructions for thedaviddias/souls-directory, covering claude.md — project context for ai assistants, project overview, tech stack, critical rules and database is convex.
nextarter-tailwind AGENTS.md
AGENTS.md instructions for agustinusnathaniel/nextarter-tailwind, covering next.js: always read docs before coding, ai agent guidance: nextarter-tailwind, 1. mental model for agents, implementation delegation and tooling strategy.
agentgram CLAUDE.md
Claude Code instructions for agentgram/agentgram, covering claude.md — agentgram development guide, project overview, tech stack, monorepo structure and core architecture patterns.