OpenAgentIDDocumentation
OpenAgentIDDocumentation

Documentation for agents

Read task-scoped Markdown, published package manifests, and the read-only documentation MCP service.

Use a specific task page and its package contract before loading the whole corpus. Human pages and machine-readable pages are generated from the same content tree.

FormatEndpointUse
Discovery/llms.txtPage titles, descriptions, and canonical URLs
Full corpus/llms-full.txtWhole-site context when required
Page MarkdownAppend .md to a page pathSmall, task-scoped context
Skill/skill.mdDocumentation workflow and boundaries
Page manifest/reference/index.jsonMachine-readable page discovery
Published API/reference/current-api.jsonExact versions, archive integrity, entry points and declarations
Package coverage/reference/coverage.jsonPublic export coverage and page mapping
MCP/api/mcpRead-only page listing, retrieval and search

MCP connection

Connect a Streamable HTTP MCP client to https://docs.openagent.id/api/mcp. The documentation service provides list_pages, get_page, and search. It only retrieves documentation. It does not generate identity keys, submit transactions, request credentials, or invoke product APIs.

Use the published contract

  • Match the exact package, version, import path and runtime conditions.
  • Respect peer dependency ranges and the limitations on each package page.
  • Treat an unverified registry release as source installation, not a public package promise.
  • Use current SDK verification for new applications; older adapters are not automatically compatible.
  • Configure identity, credential, and authority providers explicitly.
  • Never infer cross-language encryption-key restoration from ACT vectors.
  • Do not include secrets, seeds, sessions, private keys, or bearer credentials in documentation searches.

Begin with the public starter, current CLI/SDK tools, identity verification or live avatars. Use the current API reference for declaration details. Signature fences are reference declarations; the runnable examples are complete programs. The old extracted source snapshots have been removed from the live corpus.

On this page