{"slug":"denissergeevitch-agents-best-practices","source_name":"denissergeevitch/agents-best-practices","name":"Denissergeevitch/Agents Best Practices","description":"Use this skill when designing, generating an MVP blueprint for, auditing, refactoring, or explaining an agentic harness for any domain. Covers provider-neutral agent architecture for OpenAI, Anthropic, and OpenAI-compatible APIs: agent loops, tool design, permissions, system prompts, planning, goals, context compaction, memory, skills, MCP/external connectors, observability, evals, prompt caching, agent-legible environments, feedback loops, and safety.","version":2,"lift":{"pass_rate_delta_pts":31.82,"pass_rate_pct":95.5,"total_cases":22,"passed_cases":21,"tokens_delta_pct":129.6,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-13T09:10:24.102435+00:00"},"skill_score":0.9545,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":31.82,"with_pass_pct":95.5,"without_pass_pct":63.6,"tokens_delta_pct":129.6,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":22,"verdict":"mixed","never_hurt":true,"completed_at":"2026-08-13T09:10:24.102435+00:00","run_id":"d5f628cd-f4a3-4226-aa64-884156161584","version_number":2,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"b0a4d4b69e978b206bb2f1c34a1d72ccaa51441b247d3d22d7e9e37ca929e66f","raw_url":"https://app.decimal.ai/s/denissergeevitch-agents-best-practices/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/denissergeevitch-agents-best-practices"}