BioFlow

AI & agent tooling

Everything an AI assistant needs to build on BioFlow: llms.txt, markdown twins of every docs page, the OpenAPI 3.1 spec, the MCP server, and an agent prompt.

Every page of the BioFlow developer surface has a machine-readable counterpart — plain-text indexes, markdown twins, a downloadable spec, and a remote MCP server. Point your assistant at any of them.

Machine-readable surfaces

llms.txt & llms-full.txt

https://getbioflow.com/llms.txt is the compact site index (llmstxt.org convention). https://getbioflow.com/llms-full.txt goes further: the entire API reference — every operation, every error code — as one plain-text document an agent can ingest in a single fetch.

Markdown twins

Append .md to any developer page URL and you get the same content as clean markdown:

https://getbioflow.com/docs/api.md
https://getbioflow.com/docs/api/webhooks.md
https://getbioflow.com/docs/api/errors/rate-limited.md
https://getbioflow.com/docs/ai.md
https://getbioflow.com/developers.md · https://getbioflow.com/changelog.md

OpenAPI 3.1 spec

The exact document the /v1 surface is generated from — every operation, schema, scope, and quota-cost annotation:

https://getbioflow.com/docs/api/openapi.json

MCP server

For agents that should operate a workspace rather than call REST: a remote MCP server with 12 tools behind OAuth 2.1 at https://app.getbioflow.com/api/mcp. The setup guide — per-client steps, tools, scopes, discovery endpoints and denial codes — is Connecting AI assistants. Connect from Claude Code:

claude mcp add --transport http bioflow https://app.getbioflow.com/api/mcp

Agent prompt

A ready-made prompt that points the agent at the markdown docs and sets a safe key-handling ground rule:

You're helping me build on the BioFlow API. Read https://getbioflow.com/docs/api.md for the full surface (auth, quotas, errors, all 22 operations), then help me write code against https://app.getbioflow.com/v1. My API key is in the BIOFLOW_API_KEY env var — never print it.

REST docs live on the API docs hub; the human-facing platform overview is https://getbioflow.com/developers.

On this page