Borrowing it
Nothing to install: this file belongs to max-listov/stitchkit. 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/max-listov/stitchkit/master/AGENTS.mdgit clone --depth 1 https://github.com/max-listov/stitchkitWrote 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/max-listov/stitchkit/agents-md)<a href="https://agentmods.dev/instructions/max-listov/stitchkit/agents-md"><img src="https://agentmods.dev/badge/instructions/max-listov/stitchkit/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/max-listov/stitchkit/agents-md"><img src="https://agentmods.dev/badge/instructions/max-listov/stitchkit/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 | $0.03595 | $0.03595 |
| Opus 5.5 | $0.01438 | $0.01438 |
| Sonnet 5.5 | $0.00719 | $0.00719 |
| Haiku 4.5 | $0.00360 | $0.00360 |
Grade A, and why
stitchkit 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.
How it starts
The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
stitchkit — agent guide
Contract-first backend framework for Bun and Node. One defineContract() → an HTTP API, MCP tools,
AI-agent tools, a CLI and a typed client. What it is and is not: docs/PRINCIPLES.md.
📦 Building an app with stitchkit? Not this file: see the README, the guide, and in your project
node_modules/stitchkit/llms.txt(it names thellms/<entrypoint>.txtslice to load per import) orskills/stitchkitcopied into.claude/skills/.🔧 Developing stitchkit? One rule per line, each linked to where its reasoning lives: ADRs in
docs/decisions/, the rest indocs/architecture/. Setup, commands, hooks, local development against an app, PRs:CONTRIBUTING.md.
Before you change code
- Name the invariant (I1–I15) your change serves; nothing under "Not in stitchkit" ships. → PRINCIPLES
- NEVER ship a competing WebSocket or hook engine — wrap Socket.IO (
createSocketIOClient/createSocketIOServer) andreact-query-kit(createCursorQuery). → ADR 0008 - ALWAYS Zod-first: the schema is the source of truth, types come from
z.infer; never hand-write a duplicate type. → PRINCIPLES I3 - NEVER write
asin business logic. Casts live only at named boundaries —internal/typed.ts,executableAgentRuntimeToolsintools/runtime-tool.ts, adapters over untyped emitters (Socket.IO, event bus, cache bridge), the generic bridges inbrowser/client.ts— each with a comment saying why; elsewhere fix the types upstream. → ADR 0003, engineering-rules - Transport and hooks use
RuntimeContext(loose), handlersHandlerContext(typed); never cast between them. → ADR 0003 - ALWAYS keep the core Web Fetch-clean:
createHandlertakesHandlerConfig(no Bun types); Bun APIs live only increateServerandstitchkit/server. → ADR 0013 - ALWAYS keep the core generic — no domain model; scopes are free strings, no billing,
sourceis transport-only. → ADR 0002 - NEVER make the project declaration a condition of a build, test, start path or check;
stitchkit/declarationstays a leaf (packages/core/tests/project-declaration.test.ts). → ADR 0104 - ALWAYS make a declared option load-bearing and register the test that proves it in
packages/core/tests/option-effects.test.ts. → I9, engineering-rules - ALWAYS extend the owner that already exists: one implementation per job, one owner per public name, no alias, shim or parallel path. → ADR 0199, I8
- NEVER import
agent-runtime/from the core; the runtime may import the core, shared machinery goes to a neutral part (src/durability/). Every direction between parts ofsrc/is declared inpackages/core/tests/import-graph.test.ts— declare a new one there. → ADR 0197 - NEVER publish a runtime internal because it looks useful: it leaves
agent-runtimeonly when it needs no store and no run protocol, already exists proven by tests, and is typed against what the caller holds. → ADR 0142 - ALWAYS register every process a test starts with
packages/core/tests/support/process-reaper.tsbefore it can exist;bun run testfails the run, naming the survivor, when one is left. → ADR 0249 - Keep functions ≤200 and files ≤500 lines (
packages/core/tests/code-size.test.ts); an exception is a reasoned entry inpackages/core/tests/fixtures/code-size-exceptions.json, and the list only shrinks. → ADR 0199 - Put an endpoint's tool options in the one
toolgroup (tool: { name, view, ui, annotations, mcp });exposestays outside it. → ADR 0196
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 Changed · +5 tokens per session 70e7df6ee005
- 20d ago Changed · +3 lines · +85 tokens per session c7c2116fc5b6
- 1mo ago Changed · +6 lines · +154 tokens per session 0ec0a688175f
- 1mo ago Changed · +10 lines · +222 tokens per session e58b1052f37e
- 1mo ago Changed · +2 lines · +28 tokens per session f4799aab5846
- 1mo ago Changed · +1 lines · +27 tokens per session f92538e0e867
- 1mo ago Changed · -16 tokens per session 6e39ce7cc995
- 1mo ago Changed · -285 lines · -3,599 tokens per session 40cdee01301f
- 1mo ago Changed · +13 lines · +229 tokens per session cc5b15de54d3
- 1mo ago First seen · 423 lines · 6,460 tokens per session scan A d59070731197
stitchkit AGENTS.md is an instructions file published in the GitHub repository max-listov/stitchkit (3 stars, last pushed 2d ago), licensed MIT. It adds 3,595 tokens to every session, about $0.0144 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.
stitchkit AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| stitchkit AGENTS.mdmax-listov/stitchkit | Instructions file | 3 | 2d ago | MIT | A | 3,595 | 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 |