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/duthaho/claudekit/agents-mdgit clone --depth 1 https://github.com/duthaho/claudekitWrote 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/duthaho/claudekit/agents-md)<a href="https://agentmods.dev/instructions/duthaho/claudekit/agents-md"><img src="https://agentmods.dev/badge/instructions/duthaho/claudekit/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 | $0.00916 | $0.00916 |
| Opus 5 | $0.00458 | $0.00458 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00092 | $0.00092 |
Grade A, and why
claudekit 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
What this is
claudekit — a Claude Code plugin (not an app): 15 skills, 8 agents, 5 output
styles, and 3 hook scripts, distributed via a plugin marketplace. Almost everything
is markdown consumed by Claude Code at runtime; the only buildable code is the
docs website (website/, Astro Starlight). The only repo-level check is
scripts/validate-plugin.cjs (skill anatomy + version sync).
Build, test, run
- Plugin: nothing to build. Lint before committing:
node scripts/validate-plugin.cjs— validates every SKILL.md against the 8-section anatomy and plugin/marketplace version sync;node scripts/token-report.cjs --check— enforces token budgets on skill/agent frontmatter;node scripts/test-verify-evidence.cjs— exercises the evidence gate (scripts/verify-evidence.cjs: citation resolution, fake-green tripwires, and--rerunclaim-vs-actual test diffing). All three run in CI,.github/workflows/validate.yml. - Test by installing locally:
/plugin marketplace add <path-or-repo>→/plugin install claudekit→/claudekit:init - Website (
cd website/):npm install,npm run dev,npm run build,npm run preview. Deployed to Cloudflare viawrangler.jsonc(assets from./dist).
Architecture
.claude-plugin/plugin.json— plugin manifest;.claude-plugin/marketplace.json— marketplace listing. Version lives in both files; bump them together.skills/<name>/SKILL.md— one dir per skill. Onlyskills/init/has extras:templates/(hooks.json, mcp-servers.json, rules/*.md) that init scaffolds into a user's.claude/dir.agents/<name>.md— YAML frontmatter (name,descriptionwith embedded<example>blocks as an escaped single-line string,tools,memory: project)- system prompt body. See
agents/tester.md.
- system prompt body. See
output-styles/<name>.md— frontmatter must keepkeep-coding-instructions: true.scripts/*.cjs— Node hook scripts (PostToolUse etc.); they fail open — errors are swallowed by design (seescripts/auto-format.cjs).website/src/content/docs/— docs pages;reference/{skills,agents,output-styles,mcp-servers}.mdmirror the plugin contents.
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 · 59 lines · 916 tokens per session scan A 37245d654f16
claudekit AGENTS.md is an instructions file published in the GitHub repository duthaho/claudekit (97 stars, last pushed 1mo ago), licensed MIT. It adds 916 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 2026-08-30.
Other instructions, from other repositories
lynxprompt-vscode CLAUDE.md
Claude Code instructions for GeiserX/lynxprompt-vscode, covering claude.md — lynxprompt vs code, tech stack, development, architecture and key commands.
tidyfactor-design AGENTS.md
Instructions for alwkala/tidyfactor-design, covering ⚡ skill & 24 modular slash commands (7 lifecycle stages) and critical architecture (non-negotiable).
claude-code-second-brain AGENTS.md
Instructions for shofer-dev/claude-code-second-brain, covering agents.md, read first, invariants and working here.
loadout CLAUDE.md
Instructions for AmElmo/loadout, covering loadout, project structure, tech stack, key commands and test fixtures.
pixcode AGENTS.md
AGENTS.md instructions for alicomert/pixcode, covering agents.md, commands, smoke tests (manual, no npm script), architecture boundaries and agent adapters.
Warden-AI GEMINI.md
Gemini CLI instructions for rynald0cst0ltziam/Warden-AI, covering warden — context governance and verification layer, session start (important — do this first), layer 1: before starting work — context selection, layer 2: during work — tool output pruning and enforcement hooks (automatic).