Coding agents

Beignet's conventions are enforced by tooling — beignet lint checks dependency direction, beignet doctor reports registration and structure drift, and generators produce canonical output. That makes a Beignet app unusually legible to coding agents: an agent can generate an artifact, check its own work, and fix drift without guessing at project conventions. Three integration points connect agents to that tooling.

The scaffolded guide files

beignet create writes AGENTS.md and CLAUDE.md at the app root. Agents that follow the AGENTS.md convention read AGENTS.md automatically, and Claude reads CLAUDE.md; both files contain the same conventions and also apply to humans.

It carries the conventions an agent cannot discover by reading code:

Registration is not automatic. A route group, schedule, task, or outbox registry entry must be registered in its central server/ file before it runs. The guide lists what to register where, and notes that beignet doctor --fix repairs most registration drift.
Prefer generators. beignet make output lands in the right place, wires registries, and passes lint and doctor; hand-written files often miss a wiring step.
The validation loop. Run the app's Biome lint, beignet lint, beignet doctor --strict, tests, and typecheck after each change, and treat findings as the next task.
Placement rules and the naming grammar. Where feature artifacts live, and how defineX, createX, and createXProvider divide the API.

The guide deliberately excludes anything discoverable from the code or covered elsewhere: the README.md owns setup and run instructions, and AGENTS.md and CLAUDE.md own conventions. Keeping them short keeps agents reading them.

The MCP server

beignet mcp runs a Model Context Protocol server over stdio, exposing the CLI's inspection and generation commands as tools:

Tool	What it does
`routes`	List the app's routes and contracts. Read-only.
`doctor`	Report framework drift as JSON diagnostics. Accepts `{ strict?: boolean }`, default `true`. Read-only.
`doctor_fix`	Apply doctor's low-risk fixes: route-group, schedule, task, and outbox registration drift. Listener drift stays report-only and must be fixed by hand.
`lint`	Report architecture and dependency-direction diagnostics. Read-only.
`make`	Run a generator with `{ artifact, name, ...options }` — the same artifact kinds as `beignet make`.

Tool outputs are exactly the JSON the CLI's --json flags produce, so anything written against beignet doctor --json or beignet lint --json reads MCP results unchanged.

Generated apps ship a .mcp.json that registers the server through the app's package runner, so Claude Code picks it up with zero configuration. Other clients, such as Cursor and VS Code, take the same command in their own config files (.cursor/mcp.json or .vscode/mcp.json):

json

{
  "mcpServers": {
    "beignet": {
      "command": "bunx",
      "args": ["beignet", "mcp"]
    }
  }
}

In npm, pnpm, or yarn apps, use npx as the command instead. Both runners resolve the app-local beignet bin that generated apps install, so the server always matches the app's CLI version.

llms.txt

The docs site publishes two plain-text views of itself, rebuilt on every docs build:

https://beignetjs.com/llms.txt — an llms.txt-style index mirroring the docs navigation: every page with its title, URL, and description.
https://beignetjs.com/llms-full.txt — the full text of every docs page in one file.

Use the index when an agent should navigate to the right page and fetch it; use the full file when an agent retrieves grep-style over the whole corpus.

See the CLI reference for the beignet mcp command and App architecture for the structure the conventions in AGENTS.md and CLAUDE.md describe.