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/devstride/claude-plugin/agents-mdgit clone --depth 1 https://github.com/devstride/claude-pluginWhat 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 | $0.00563 | $0.00563 |
| Opus 5 | $0.00282 | $0.00282 |
| Sonnet 5 | $0.00113 | $0.00113 |
| Haiku 4.5 | $0.00056 | $0.00056 |
Grade A, and why
claude-plugin 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 yesterday.
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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working in this repository
This repo is public. Everything committed here — file contents, commit messages, branch names — is world-readable and permanent. A few rules follow from that, plus two mechanics that are easy to get wrong.
Public-repo hygiene
- No internal references. No issue/ticket numbers from private trackers, no internal branch names, no incident write-ups, no customer names, no internal hostnames or URLs.
- No session or tooling trailers in commit messages. Ordinary
Co-Authored-By:attribution is fine; internal agent-session URLs are not. - Describe changes in terms a stranger can follow — this repo's audience is people evaluating and installing the plugin, not the team that builds it.
Releasing: the version field is the update signal
version in .claude-plugin/plugin.json is both the pin and the cache key. Installed users
keep a cached copy at that version, so pushing new commits without bumping version means
existing installs never receive the change. Every user-visible change ships with a bump.
Tag releases with claude plugin tag, which creates a {name}--v{version} git tag and checks
that plugin.json and the marketplace entry agree.
Validation caveat
claude plugin validate <repo-root> validates the marketplace manifest only when a
marketplace.json is present — it silently never checks plugin.json. Validate the plugin
manifest explicitly (point the command at a copy without the marketplace file, or validate both
paths individually) rather than trusting a single root-level run.
Layout
-
References are FLAT under
skills/<name>/references/— the needle corpus globskills/*/references/*.mdis one level deep, so a nested directory is invisible to the rule-loss check. (The standing exception,skills/setup/references/docs-skill-templates/, holds templates, not rules.) -
.claude-plugin/holdsplugin.jsonandmarketplace.json— and nothing else. -
Component directories live at the repo root:
skills/is auto-scanned, so it does not need declaring inplugin.json. -
Shared reference docs belong to the skill that owns them (
skills/<name>/references/), not to a repo-level directory. -
scripts/holds repo-level maintenance scripts (validate.sh,measure-cost.sh, their tests). Nothing at runtime reads them and the plugin loader ignores the directory — it is not a component directory, and it must not move under.claude-plugin/.
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.
- yesterday First seen · 47 lines · 563 tokens per session scan A fef455e72338
claude-plugin AGENTS.md is an instructions file published in the GitHub repository devstride/claude-plugin (2 stars, last pushed 3d ago), licensed MIT. It adds 563 tokens to every session, about $0.0028 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
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.