{"version":"0.56.1","date":"2026-01-28","dateFormatted":"January 28, 2026","url":"https://docs.factory.ai/changelog/release-notes#v0.56.1","jsonUrl":"https://docs.factory.ai/changelog/releases/0.56.1.json","summary":"MCP tool management and Read tool fixes","surfaces":["cli"],"changes":[{"type":"improvement","surface":"cli","title":"MCP tool management"},{"type":"fix","surface":"cli","title":"Read tool images"},{"type":"fix","surface":"cli","title":"Cache block limits"}],"security":[],"deprecations":[],"contentText":"Improvements\n\n- MCP tool management - Improved MCP tool status tracking to prevent cross-contamination between servers\n\nBug fixes\n\n- Read tool images - Fixed image support in the Read tool, along with rate limit retry and reasoning fixes\n- Cache block limits - Ensured no more than 3 cache blocks are sent to the LLM to prevent context issues"}