Borrowing it
Nothing to install: this file belongs to pachca/openapi. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/pachca/openapi/main/AGENTS.mdgit clone --depth 1 https://github.com/pachca/openapiWrote 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/pachca/openapi/agents-md)<a href="https://agentmods.dev/instructions/pachca/openapi/agents-md"><img src="https://agentmods.dev/badge/instructions/pachca/openapi/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/pachca/openapi/agents-md"><img src="https://agentmods.dev/badge/instructions/pachca/openapi/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01795 | $0.01795 |
| Opus 5.5 | $0.00718 | $0.00718 |
| Sonnet 5.5 | $0.00359 | $0.00359 |
| Haiku 4.5 | $0.00179 | $0.00179 |
Grade A, and why
openapi 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI coding agents working in this repository. (Using the Pachca
API instead of contributing here? See https://dev.pachca.com/skill.md,
https://dev.pachca.com/llms.txt, the MCP server https://mcp.pachca.com/mcp,
or the plugin in plugins/pachca/ — not this file.)
What this repo is
Monorepo (Turborepo + bun) for the Pachca API platform. TypeSpec is the single source of truth; everything else is generated from it.
| Path | What |
|---|---|
packages/spec |
TypeSpec (typespec.tsp) → openapi.yaml; English overlay → openapi.en.yaml; MCP tool manifest (mcp-manifest.json), examples |
packages/cli |
oclif CLI @pachca/cli, generated from OpenAPI |
packages/generator |
Multi-language SDK generator |
packages/openapi-parser |
Shared OpenAPI parser |
packages/mcp-manifest |
npm package @pachca/mcp-manifest: the MCP tool manifest and its JSON Schema, copied from packages/spec by the build |
apps/docs |
Next.js 16 docs site (dev.pachca.com): content/ MDX, lib/, scripts/ generators, public/ generated artifacts |
plugins/pachca |
The Pachca plugin for Claude Code, the Claude app, Codex, VS Code and GitHub Copilot, and Cursor: hand-written skills, MCP configs and one manifest per host |
integrations/n8n |
n8n community node |
Build & check
npx turbo build # compiles TypeSpec + builds all packages + regenerates generated files
npx turbo check # lint + typecheck + knip + format:check + test — MUST pass before commit (CI runs this)
bun turbo dev # docs dev server (Next.js 16)
Standard loop: edit source → npx turbo build → npx turbo check →
review generated diffs → commit. Branch from origin/main; open a PR
(never push to main).
Do NOT hand-edit generated files — edit the generator
| Generated | Source of truth |
|---|---|
packages/spec/openapi.yaml |
packages/spec/typespec.tsp (then npx turbo build --filter=@pachca/spec --force) |
packages/spec/openapi.en.yaml |
packages/spec/overlay.en.yaml + scripts/apply-overlay.ts (a changed Russian text also needs bun run overlay:fingerprints in packages/spec) |
packages/spec/mcp-manifest.json |
the spec + packages/spec/mcp-core.ts, mcp-tools.ts, scripts/generate-mcp-manifest.ts (then bun run generate-mcp in packages/spec) |
packages/mcp-manifest/mcp-manifest.json, mcp-manifest.schema.json |
copies of packages/spec/mcp-manifest.json and mcp-manifest.schema.json (npx turbo build --filter=@pachca/mcp-manifest); a changed manifest needs a new version on top of packages/mcp-manifest/CHANGELOG.md and in its package.json, and a release entry with product mcp and that version in apps/docs/data/releases.json (the package itself is not there) |
apps/docs/public/llms*.txt, public/**/*.md, Postman collection |
apps/docs/scripts/generate-llms.ts |
apps/docs/public/skill.md, public/.well-known/**, public/skills/*.zip, plugins/pachca/evals/mocks/pachca/_tools.json, and the version in plugins/pachca/plugin.json, .codex-plugin/plugin.json, .claude-plugin/plugin.json, .cursor-plugin/plugin.json (recorded in apps/docs/scripts/skills/plugin-version.json) |
the hand-written plugin in plugins/pachca/, checked and published by apps/docs/scripts/skills/generate.ts (links, MCP tool and CLI command names, size, manifests in step; the version is one patch above the one in main once the plugin files change) |
packages/cli/src/commands/** (except auth/config), packages/cli/CHANGELOG.md |
packages/cli/scripts/generate-cli.ts; src/data/changelog.json |
| n8n node files | integrations/n8n/scripts/generate-n8n.ts |
AGENTS.md (this file) |
apps/docs/scripts/skills/generate.ts (generateAgentsMd()) |
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 Changed · -7 tokens per session e8d4a1caff48
- 5d ago Changed · +34 tokens per session 350f675f900f
- 8d ago Changed · +4 lines · +287 tokens per session e0113f1e3ecd
- 17d ago Changed · +4 lines · +260 tokens per session ef1c990a11b7
- 2mo ago First seen · 64 lines · 1,221 tokens per session scan A 03d15746eae4
openapi AGENTS.md is an instructions file published in the GitHub repository pachca/openapi (11 stars, last pushed 2d ago), licensed MIT. It adds 1,795 tokens to every session, about $0.0072 per session on Opus 5.5. 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.
openapi AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| openapi AGENTS.mdpachca/openapi | Instructions file | 11 | 2d ago | MIT | A | 1,795 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |