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/pbnz/repo-kit/agents-mdgit clone --depth 1 https://github.com/PBNZ/repo-kitWrote 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/pbnz/repo-kit/agents-md)<a href="https://agentmods.dev/instructions/pbnz/repo-kit/agents-md"><img src="https://agentmods.dev/badge/instructions/pbnz/repo-kit/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 | $0.00648 | $0.00648 |
| Opus 5 | $0.00324 | $0.00324 |
| Sonnet 5 | $0.00130 | $0.00130 |
| Haiku 4.5 | $0.00065 | $0.00065 |
Grade A, and why
repo-kit 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — RepoKit
RepoKit is a reusable GitHub-repository standard plus a /new-repo scaffolder, packaged as a
Claude Code plugin. It stamps consistent, navigable repos from commit #1, organised by
type × tier so private repos stay light and public/published ones get the right ceremony.
This repo dogfoods its own standard (Core + Public tier).
START-HERE map — where things live
| You want… | Look in |
|---|---|
| What RepoKit is / how to install | README.md |
| Where work was left (resume state / checkpoint) | docs/CHECKPOINT.md |
| The standard itself (tiers, where-things-live, do/don't) | plugins/repokit/skills/repo-standard/standard/the-standard.md |
| Commit / changelog / version conventions | plugins/repokit/skills/repo-standard/standard/commit-conventions.md |
| Pre-commit / pre-PR / session-end checklists | plugins/repokit/skills/repo-standard/standard/pre-commit-checklist.md, pre-pr-checklist.md, session-end-checklist.md |
| Doc formatting & the living-docs pattern | plugins/repokit/skills/repo-standard/standard/doc-style.md, living-docs.md |
| What to test per repo type | plugins/repokit/skills/repo-standard/standard/testing-matrix.md |
The scaffolding methodology (/new-repo) |
plugins/repokit/skills/new-repo/SKILL.md |
| Templates stamped into new repos | plugins/repokit/skills/new-repo/templates/ |
| Decisions & rationale (learnings log) | docs/adr/ |
| Validation scripts (run before committing) | scripts/ + .github/workflows/validate.yml |
| Standard-compliance self-check (CI dogfoods it here) | plugins/repokit/skills/new-repo/templates/core/scripts/repokit-check.ps1 |
Ground rules
AGENTS.mdis the canonical agent file;CLAUDE.mdis a thin@AGENTS.mdimport (Claude Code readsCLAUDE.md, notAGENTS.md).- Default to verified, simple, proven practices. Conventional Commits; one concern per PR.
- When working in this repo, the
repo-standardskill auto-loads the conventions and the pre-commit / pre-PR checklists — follow them before committing or opening a PR. - Templates and
SKILL.mdfiles are prompt content — keep them plain prose, no secrets, no email addresses (CI enforces this viascripts/check_*.py). - RepoKit publishes to its own marketplace; "publish" = bump versions in
plugin.json/marketplace.json/CHANGELOG.md, then tag and push. Seedocs/adr/0001-repokit-architecture.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 · 37 lines · 648 tokens per session scan A b645ede28ed8
repo-kit AGENTS.md is an instructions file published in the GitHub repository PBNZ/repo-kit (1 stars, last pushed 22d ago), licensed Apache-2.0. It adds 648 tokens to every session, about $0.0032 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.
adr copilot-instructions.md
Instructions for zircote-plugins/adr, covering copilot instructions, project overview, key components, plugin structure and development guidelines.
docflow AGENTS.md
Instructions for EvolveHQ/docflow, covering agents.md, what this repository is, repository structure, hard rules when editing adrs and domain-specific hard rules.
code-decisions CLAUDE.md
Instructions for zimalabs/code-decisions, covering claude.md, what this is, commands, key concepts and storage.
agent-atelier CLAUDE.md
Instructions for big-emotion/agent-atelier, covering claude.md, what this repo is, registry layout, commands and single test.
ship-it AGENTS.md
Instructions for khrichtchatyi/ship-it: Each rule or paragraph in a SKILL.md file must be a single sentence; use semicolons, colons, or em dashes to chain related clauses inside one sentence, but never start a second sentence inside the same paragraph — split it into a new paragraph instead.