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/underpass-ai/kmp/agents-mdgit clone --depth 1 https://github.com/underpass-ai/kmpWhat 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 | $0.00416 | $0.00416 |
| Opus 5 | $0.00208 | $0.00208 |
| Sonnet 5 | $0.00083 | $0.00083 |
| Haiku 4.5 | $0.00042 | $0.00042 |
Grade A, and why
kmp 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 2d 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.
What it actually says
KMP repository working agreement
Public README parity
KMP has three public README surfaces and a change to the product overview must update all three in the same change:
README.md— GitHub repository;plugins/kmp/README.md— Codex and Claude plugin marketplaces;crates/kmp-mcp/README.md— crates.io.
The marked kmp:public-overview block in plugins/kmp/README.md is canonical.
It carries the common product contract: local-first SQLite memory, decisions
and evidence rather than transcripts, Codex and Claude Code, and the ten
memory plus three semantic view tools over shared ChronoLoom. After editing
it, run:
cargo run --locked --quiet -p kmp-release -- readme sync
Do not hand-edit the corresponding marked blocks in the other two files. Release preparation synchronizes them and CI rejects both byte drift and a missing product-contract claim, but the generated changes must still be reviewed and committed together.
Guide parity
The public READMEs explain the product; the installed guides teach it. A
change to KMP's tools, verb semantics, storage model, clocks, viewer or setup
must update plugins/kmp/guide/editorial.json in the same change and regenerate
both guide abouts with:
cargo run --locked --quiet -p kmp-release -- guide assets write --binary target/debug/kmp-mcp
guide:kmp-agent is the exact operational guide for agents. guide:kmp is a
shorter human path opened visually through open:guide; do not collapse them
into one audience. Setup, update, the three public READMEs and both guides must
describe the same shipped version before a release is prepared. The supported
scripts/release.sh version verb builds the bumped engine and regenerates both
guide assets; candidate and tag paths reject a stale guide envelope.
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.
- 2d ago First seen · 44 lines · 416 tokens per session scan A a47615e31114
kmp AGENTS.md is an instructions file published in the GitHub repository underpass-ai/kmp (0 stars, last pushed 2d ago), licensed Apache-2.0. It adds 416 tokens to every session, about $0.0021 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
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).
spec-kit AGENTS.md
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.
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.
langchain AGENTS.md
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.
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).
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.