Machine-readable docs

Point any agent at these resources to read and learn Valyd by plain fetch — the lightweight alternative to a docs MCP server, with nothing to run. Everything here is same-origin and served as static files.

A compact, link-first index of the whole documentation set. Start here: fetch it, then follow the per-page .md mirrors it links to.

llms.txt — the docs index

Every page concatenated into one file for whole-corpus ingestion or a single retrieval pass.

llms-full.txt — the full corpus

OpenAPI 3.1 for the Verification API (/api/v2/*): sessions, Unique Human API checks, workflows, and webhooks.

Verification OpenAPI

OpenAPI 3.1 for Connect with Valyd (OAuth 2.0 / OIDC) and the account APIs.

Connect OpenAPI

A machine-readable summary of what Valyd can do — the capabilities an agent can discover and map to the docs above.

Capability manifest

The human-readable companion: how an agent should reason about login vs verification vs data ownership before writing code.

Agent integration guide

The resources in one table

Resource	URL	What it is
Docs index	`/llms.txt`	Link-first index of every page.
Full corpus	`/llms-full.txt`	Every page concatenated into one file.
Verification OpenAPI	`/openapi/valyd-verify.json`	OpenAPI 3.1 for the Verification API.
Connect OpenAPI	`/openapi/valyd-id.json`	OpenAPI 3.1 for Connect with Valyd.
Capability manifest	`/agent/capabilities.json`	Discoverable capability summary.
Agent guide	`/ai/agent-guide`	How to integrate as an agent.

Per-page .md mirrors are linked from llms.txt — every doc page is also fetchable as plain Markdown at its own .md URL. Prefer callable tools over reading? See the MCP server quickstart.