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/bastndev/skills/agents-mdgit clone --depth 1 https://github.com/bastndev/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/bastndev/skills/agents-md)<a href="https://agentmods.dev/instructions/bastndev/skills/agents-md"><img src="https://agentmods.dev/badge/instructions/bastndev/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 | $0.00738 | $0.00738 |
| Opus 5 | $0.00369 | $0.00369 |
| Sonnet 5 | $0.00148 | $0.00148 |
| Haiku 4.5 | $0.00074 | $0.00074 |
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 3d 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
What this repo is
A collection of skill definitions for the OpenCode CLI (npx skills add bastndev/skills). There is no build step, no package manager at root, no TypeScript to compile, and no tests to run. The deliverable is the Markdown content itself.
Directory layout
skills/<name>/SKILL.md— The actual skill definition loaded at runtime. This is the most important file in the repo. Edits here directly change user-facing behavior.skills/<name>/README.md— Human-facing documentation (shown on GitHub, linked from root README).skills/<name>/references/— Optional reference material bundled with a skill.skills/<name>/scripts/— Optional helper scripts (e.g.l10n-sync).test/— Historical test outputs and fixtures. These are not runnable tests; they are reference snapshots from past skill executions.public/docs/— Translated root README files (11 languages). Maintained byl10n-sync.CLAUDE.md— Single line:@AGENTS.md. Acts as a pointer so Claude Code loads this file.
Skill file conventions
Every SKILL.md starts with YAML frontmatter:
---
name: skill-name
description: "One-line description used by the skill registry."
license: Complete terms in LICENSE.txt
metadata:
author: bastndev
version: "x.y.z"
---
After the frontmatter, the body defines the skill's operating rules, report formats, and phase-by-phase instructions. These are prose instructions for AI agents, not executable code.
Key facts for editing
- No lint/typecheck/build commands exist. The repo is pure Markdown.
- Prettier is configured but ignores
skills/. Only root files andtest/fixtures are formatted. Runnpx prettier --check .at root to verify formatting on those files. .gitignoreexcludes.vscode/,NOTES.md, andtest/test.md. Don't un-ignore these.test/start-skills/test-astro/contains 12 Astro project fixtures generated bystart-astro. These are reference outputs, not something tonpm installor run.test/end/andtest/middle/contain numbered result snapshots (.txt/.md). These are historical records of past skill runs.
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.
- 3d ago First seen · 57 lines · 738 tokens per session scan A 660caf921bcb
skills AGENTS.md is an instructions file published in the GitHub repository bastndev/skills (2 stars, last pushed 25d ago), licensed MIT. It adds 738 tokens to every session, about $0.0037 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
agentistics CLAUDE.md
Claude Code instructions for blpsoares/agentistics, covering agentistics — claude.md, process rules live in agents.md, not here, language convention, monorepo structure and architecture.
agentistics AGENTS.md
AGENTS.md instructions for blpsoares/agentistics, covering before starting implementation work, opening a pr, documentation, scope discipline and discussions.
uk-legal-plugins CLAUDE.md
Claude Code instructions for uk-agents/uk-legal-plugins, covering claude.md, target ecosystems, documentation index, layout and validation — run before opening a pr.
agents-md AGENTS.md
AGENTS.md instructions for Austin1serb/agents-md, covering coding agent instructions, operating principles, context discipline, command output and good byte capping examples.
polyspace-agentic-toolkit AGENTS.md
AGENTS.md instructions for mathworks/polyspace-agentic-toolkit, covering polyspace agentic toolkit — agent instructions, setup, skills and mcp tools.
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).