{"schema_version":"newruntime-agent-readable-v0.2","type":"post","stable_id":"post:genkit-agent-skills-progressive-disclosure","slug":"genkit-agent-skills-progressive-disclosure","title":"Genkit Adds Progressive Disclosure For Agent Skills","description":"Genkit now loads Agent Skills through middleware that discovers SKILL.md metadata first and activates full instructions, references, and scripts only when needed.","retrieval_nugget":"Genkit now loads Agent Skills through middleware that discovers SKILL.md metadata first and activates full instructions, references, and scripts only when needed. Google added Agent Skills support to Genkit across TypeScript, Go, Dart, and Python. The implementation makes progressive disclosure a harness behavior rather than a prompting convention. A skill remains a normal directory.","status":"published","published_at":"2026-08-01","updated_at":"2026-08-01","record_date":"2026-08-01","date_kind":"published_at","topics":["skills","context-engineering","genkit","agent-harnesses"],"source_urls":["https://developers.googleblog.com/enable-on-demand-expertise-with-agent-skills-in-genkit-go"],"visuals":[{"id":"genkit-agent-skills-progressive-disclosure","kind":"editorial-diagram","role":"hero","src":"https://newruntime.com/images/posts/genkit-agent-skills-progressive-disclosure.webp","alt":"Hand-drawn progressive disclosure flow where skill metadata is scanned first, a matching request activates one SKILL file, and references or scripts load only for execution.","caption":"Genkit makes skill activation a middleware lifecycle instead of permanent prompt stuffing.","credit":"New Runtime synthesis from Google Developers Blog","source_url":"https://developers.googleblog.com/enable-on-demand-expertise-with-agent-skills-in-genkit-go","generated_with":"gemini-3.1-flash-image","width":1600,"height":900,"legend":[{"label":"Discovery","description":"The harness scans skill folders and exposes compact frontmatter descriptions."},{"label":"Activation","description":"A matching request triggers the full SKILL instructions through a dedicated tool."},{"label":"Execution","description":"References, assets, and scripts load only when the active workflow requires them."}]}],"telegram_message_id":2902,"telegram_url":"https://t.me/qwgai/2902","telegram_message_ids":[2902,2903],"telegram_delivery_mode":"text_then_media","telegram_media_url":"https://t.me/qwgai/2903","routes":{"html":"https://newruntime.com/posts/genkit-agent-skills-progressive-disclosure/","markdown":"https://newruntime.com/posts/genkit-agent-skills-progressive-disclosure.md","json":"https://newruntime.com/posts/genkit-agent-skills-progressive-disclosure.json"},"source_format":"markdown","next_reads":[{"type":"topic","path":"/topics/context-engineering/","reason":"Explore the context engineering topic hub.","url":"https://newruntime.com/topics/context-engineering/","title":"Context engineering - New Runtime","media_type":"text/html"},{"type":"topic","path":"/topics/skills/","reason":"Explore the skills topic hub.","url":"https://newruntime.com/topics/skills/","title":"Agent skills - New Runtime","media_type":"text/html"},{"type":"related_material","path":"/posts/chatgpt-agent-loop-efficiency-stack/","reason":"Shares agent harnesses and context engineering.","url":"https://newruntime.com/posts/chatgpt-agent-loop-efficiency-stack/","title":"ChatGPT Cuts Repeated Work Across The Agent Stack","media_type":"text/html"},{"type":"related_material","path":"/posts/anthropic-tool-search-programmatic-calls/","reason":"Shares agent harnesses and context engineering.","url":"https://newruntime.com/posts/anthropic-tool-search-programmatic-calls/","title":"Anthropic Moves Large Tool Libraries Out Of Context","media_type":"text/html"},{"type":"related_material","path":"/posts/drskill-agent-loadout-audit/","reason":"Shares context engineering and skills.","url":"https://newruntime.com/posts/drskill-agent-loadout-audit/","title":"Dr. Skill Audits What An Agent Loads Before It Works","media_type":"text/html"}]}