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/arcjet/skills/agents-mdgit clone --depth 1 https://github.com/arcjet/skillsWhat 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 | $0.00688 | $0.00688 |
| Opus 5 | $0.00344 | $0.00344 |
| Sonnet 5 | $0.00138 | $0.00138 |
| Haiku 4.5 | $0.00069 | $0.00069 |
Grade A, and why
skills 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 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 — 33 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Loose lessons learned while using the skill-creator skill (at .agents/skills/skill-creator/). Terse on purpose. Always read .agents/skills/skill-creator/SKILL.md first — this file only supplements it.
Subagent prompts for test runs
- Keep prompts minimal. The prescribed format (
Skill path / Task / Input files / Save outputs to) is enough — don't add framing like "you're running an eval," "report what you did," or environment hints the agent should discover itself (e.g. that the CLI is pre-authenticated). - Don't add restrictions like "don't run the server" or "typechecking optional." Those bias the agent away from the verification step you're trying to measure.
Aggregation gotchas
- The aggregator expects per-run subdirectories:
eval-X/<config>/run-N/grading.json(plustiming.json). If you save outputs ateval-X/<config>/grading.json(norun-N), it picks up nothing and reports 0%. - Config ordering in
benchmark.mdis alphabetical, which meansnew_skill/old_skillandwith_skill/without_skillflip the delta sign. Sanity-check the sign by hand.
Capturing subagent transcripts
- Each
Agenttask writes its JSONL transcript to/tmp/claude-*/-workspaces-*/<session>/tasks/<task_id>.output. Don'tcatorReadit — it's the full transcript and will blow your context. - Do
cpit into the run directory astranscript.jsonlonce the task completes. The viewer also looks fortranscript.mdat the same level.
Eval viewer
- It embeds every file under
outputs/into a single inline<script>. Without an exclude list it'll happily inline 482 MB ofnode_modules, and a stray backtick inside a vendored file crashes the page. The current code excludesnode_modules,__pycache__,.venv,.next,dist,build, lockfiles, etc. — extend that list if a new ecosystem shows up. - Bind to
0.0.0.0(not127.0.0.1) so devcontainer / remote setups can reach it via the forwarded port.
When the same baseline serves multiple iterations
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 · 33 lines · 688 tokens per session scan A 6504c220d8dd
skills AGENTS.md is an instructions file published in the GitHub repository arcjet/skills (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 688 tokens to every session, about $0.0034 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
skills AGENTS.md
AGENTS.md instructions for payloadcms/skills, covering payload skills repository, repository overview, creating a new skill, directory structure and naming conventions.
claude-skill-registry CLAUDE.md
Claude Code instructions for majiayu000/claude-skill-registry, covering claude.md, repo map, how to decide where to edit, generated vs owned files in main and publish flow.
agent-skills AGENTS.md
AGENTS.md instructions for tech-leads-club/agent-skills, covering agents.md, workflow orchestration, 1. plan mode default, 2. subagent strategy and 3. verification before done.
Skills CLAUDE.md
Instructions for kriscard/Skills, covering claude.md, requirements, repository structure, skill structure and frontmatter rules (verified against official docs).
harness-anchor CLAUDE.md
Instructions for Redtropig/harness-anchor, covering harness-anchor — contributor & ai collaborator guidelines, if you are an ai agent editing this plugin, design invariants (do not break), shell hazards no tool in this repo catches and authoring a new skill (when explicitly asked).
pivoshenko.ai CLAUDE.md
Instructions for pivoshenko/pivoshenko.ai, covering claude.md, what this repo is, commands, asset layout and the sync contract and frontmatter contract.