Your docs, ready for AI agents: what we shipped and what's next
Blog post from Redocly
Redocly has expanded its documentation platform to make API content more accessible and reliable for AI agents, adding a built-in Docs MCP server at `/mcp` that lets supported tools search documentation, inspect OpenAPI and GraphQL APIs, access versioned and localized content, and enforce existing RBAC controls. Recent improvements include support for the latest MCP protocol, simpler one-click client setup, automatic token refresh, session persistence, standard tool safety annotations, agent-discovery metadata, task-specific agent skills, and enhanced `llms.txt` output for code samples and complex schemas. Its Reunite AI Assistant now uses account-specific information such as build logs, configurations, plans, and membership data to provide more contextual support, while the early-access Reviewer examines pull requests for API, documentation, security, configuration, and cross-document inconsistencies. Planned features include opt-in Gateway MCP for allowing agents to call APIs, a sandboxed code mode intended to reduce token use and cost, custom MCP tools, an automated Writer agent that proposes documentation fixes, and an embeddable AI Assistant for web applications.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.