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/bitcoin21ideas/skills/agents-mdgit clone --depth 1 https://github.com/bitcoin21ideas/skillsWrote 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/bitcoin21ideas/skills/agents-md)<a href="https://agentmods.dev/instructions/bitcoin21ideas/skills/agents-md"><img src="https://agentmods.dev/badge/instructions/bitcoin21ideas/skills/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 | $0.00904 | $0.00904 |
| Opus 5 | $0.00452 | $0.00452 |
| Sonnet 5 | $0.00181 | $0.00181 |
| Haiku 4.5 | $0.00090 | $0.00090 |
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 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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI coding agents (Codex, Claude Code, Cursor, Copilot, Gemini, and others) working in this repository. This is a distribution repo for reusable agent skills — there is no application to build or run. Most contributions add or edit a skill.
For human-facing usage and install instructions, see README.md.
Repository layout
skills/<name>/
SKILL.md # required — the skill itself (the agent-facing instructions)
README.md # required — human-facing overview, install, examples
examples/ # optional — sample sessions or supporting docs
Each skill is a self-contained folder under skills/. Keep skills independent — a user should be able to take one folder without the rest.
Skill conventions
SKILL.mdfrontmatter uses YAML with:name— kebab-case, matches the folder name (this is the invocation name, e.g./commit).description— what the skill does and when to use it. Phrase triggers as explicit user actions ("when the user says…", "run when the user asks to…") so cross-agent harnesses don't auto-fire it spuriously.disable-model-invocation: true— set this for explicit-only skills (e.g.pressure-test); omit it for skills that should also auto-trigger from natural-language intent (e.g.commit, which still waits for plan approval before committing). Claude Code and Copilot honor the field; agents that don't recognize it ignore it safely and fall back to description matching.metadata:block holdingversion(semver string),author, andsourceif adapted from someone else's work. Use this nested block, not top-level keys.
- Cross-agent first. Unknown frontmatter fields are ignored per the agentskills.io standard, so Claude-specific fields are safe. Don't write a skill that only works in one agent unless it's inherently agent-specific.
- Progressive disclosure. Keep
SKILL.mdfocused; move long or rarely-needed detail into sibling files (e.g.tailoring.md,examples/) that the agent loads only when relevant. - Every promoted skill must be listed in the root
README.mdskills table, with the name linked to its per-skillREADME.md.
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 · 56 lines · 904 tokens per session scan A 180dcbaddeba
skills AGENTS.md is an instructions file published in the GitHub repository bitcoin21ideas/skills (1 stars, last pushed 2mo ago), licensed MIT. It adds 904 tokens to every session, about $0.0045 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
dev-agent-skills CLAUDE.md
Instructions for fvadicamo/dev-agent-skills, covering claude.md, project overview, architecture, key file: marketplace.json and skill anatomy.
qa-orchestra CLAUDE.md
Claude Code instructions for Anasss/qa-orchestra, covering qa orchestra — claude code instructions, what this project is, quick start, 1. clone into your workspace and 2. fill in your project context.
chealth CLAUDE.md
Instructions for danielithomas/chealth, covering claude.md, project purpose, language & tooling, build & test and architecture.
claude-goal-scaffold CLAUDE.md
Claude Code instructions for wilwaldon/claude-goal-scaffold, covering project: [your project name], stack, structure, design direction and content source.
kortex CLAUDE.md
Claude Code instructions for Heapy/kortex, a project described as: Agent Skills for Kotlin Development.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).