{"schema_version":"newruntime-agent-readable-v0.2","type":"post","stable_id":"post:drskill-agent-loadout-audit","slug":"drskill-agent-loadout-audit","title":"Dr. Skill Audits What An Agent Loads Before It Works","description":"Dr. Skill scans skills and MCP servers for collisions, duplication, secrets, drift, missing metadata, and unused loadout, with local and CI-friendly commands.","retrieval_nugget":"Dr. Skill scans skills and MCP servers for collisions, duplication, secrets, drift, missing metadata, and unused loadout, with local and CI-friendly commands. Skill libraries can degrade an agent before the first user message. Drew Breunig describes an enterprise agent loading more than 600 skills by default and a Hermes installation selecting the wrong note-taking skill from a much smaller but.","status":"published","published_at":"2026-08-01","updated_at":"2026-08-01","record_date":"2026-08-01","date_kind":"published_at","topics":["skills","mcp","developer-tools","context-engineering"],"source_urls":["https://dbreunig.com/2026/07/24/manage-your-agent-s-loadout-with-dr-skill.html"],"visuals":[{"id":"drskill-agent-loadout-audit","kind":"editorial-diagram","role":"hero","src":"https://newruntime.com/images/posts/drskill-agent-loadout-audit.webp","alt":"Hand-drawn agent loadout inspection where a scanner finds duplicate skills, overlapping descriptions, risky configuration, drift, and unused tools before a clean runtime starts.","caption":"Dr. Skill treats the set of loaded skills and MCP tools as a dependency surface that needs linting and usage evidence.","credit":"New Runtime synthesis from Drew Breunig","source_url":"https://dbreunig.com/2026/07/24/manage-your-agent-s-loadout-with-dr-skill.html","generated_with":"gemini-3.1-flash-image","width":1600,"height":900,"legend":[{"label":"Scan","description":"Static checks find missing descriptions, duplicate installs, config risks, and version drift."},{"label":"Compare","description":"An optional deep pass judges whether names, descriptions, and scopes overlap."},{"label":"Audit","description":"Runtime traces reveal which skills and tools are used and what queries activated them."}]}],"routes":{"html":"https://newruntime.com/posts/drskill-agent-loadout-audit/","markdown":"https://newruntime.com/posts/drskill-agent-loadout-audit.md","json":"https://newruntime.com/posts/drskill-agent-loadout-audit.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/developer-tools/","reason":"Explore the developer tools topic hub.","url":"https://newruntime.com/topics/developer-tools/","title":"Developer Tools - New Runtime","media_type":"text/html"},{"type":"related_material","path":"/posts/mem0-context-wipe-memory-layer/","reason":"Shares context engineering and developer tools.","url":"https://newruntime.com/posts/mem0-context-wipe-memory-layer/","title":"Mem0 Shows Agent Memory Should Not All Live in Context","media_type":"text/html"},{"type":"related_material","path":"/posts/copilotkit-react-mcp-client-interface/","reason":"Shares developer tools and mcp.","url":"https://newruntime.com/posts/copilotkit-react-mcp-client-interface/","title":"CopilotKit Brings MCP Tool Calls Into The React Interface","media_type":"text/html"},{"type":"related_material","path":"/posts/anthropic-tool-search-programmatic-calls/","reason":"Shares context engineering and mcp.","url":"https://newruntime.com/posts/anthropic-tool-search-programmatic-calls/","title":"Anthropic Moves Large Tool Libraries Out Of Context","media_type":"text/html"}]}