{"slug": "boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents", "title": "Boris Cherny's Claude Code workflow as a drop-in CLAUDE.md (workflow orchestration, subagents, self-improvement loop, verification, autonomous bug fixing). Source: thread from @bcherny.", "summary": "Boris Cherny shared a detailed Claude Code workflow designed to improve AI agent reliability through structured planning, subagent orchestration, and self-improvement loops. The methodology emphasizes plan-first execution, continuous verification, and a lessons-learned system to prevent repeated mistakes. It also includes autonomous bug fixing and CI failure resolution without user hand-holding.", "body_md": "[CLAUDE.md](https://gist.github.com/shahshrey/44da3ef7f0120297466c67bda526b018#file-claude-md)\n- Enter plan mode for ANY non-trivial task (3+ steps or architectural decisions)\n- If something goes sideways, STOP and re-plan immediately - don't keep pushing\n- Use plan mode for verification steps, not just building\n- Write detailed specs upfront to reduce ambiguity\n\n- Use subagents liberally to keep main context window clean\n- Offload research, exploration, and parallel analysis to subagents\n- For complex problems, throw more compute at it via subagents\n- One tack per subagent for focused execution\n\n- After ANY correction from the user: update\n`tasks/lessons.md`\n\nwith the pattern\n- Write rules for yourself that prevent the same mistake\n- Ruthlessly iterate on these lessons until mistake rate drops\n- Review lessons at session start for relevant project\n\n### 4. Verification Before Done\n\n- Never mark a task complete without proving it works\n- Diff behavior between main and your changes when relevant\n- Ask yourself: \"Would a staff engineer approve this?\"\n- Run tests, check logs, demonstrate correctness\n\n### 5. Demand Elegance (Balanced)\n\n- For non-trivial changes: pause and ask \"is there a more elegant way?\"\n- If a fix feels hacky: \"Knowing everything I know now, implement the elegant solution\"\n- Skip this for simple, obvious fixes - don't over-engineer\n- Challenge your own work before presenting it\n\n- When given a bug report: just fix it. Don't ask for hand-holding\n- Point at logs, errors, failing tests - then resolve them\n- Zero context switching required from the user\n- Go fix failing CI tests without being told how\n\n**Plan First**: Write plan to `tasks/todo.md`\n\nwith checkable items\n**Verify Plan**: Check in before starting implementation\n**Track Progress**: Mark items complete as you go\n**Explain Changes**: High-level summary at each step\n**Document Results**: Add review section to `tasks/todo.md`\n\n**Capture Lessons**: Update `tasks/lessons.md`\n\nafter corrections\n\n**Simplicity First**: Make every change as simple as possible. Impact minimal code.\n**No Laziness**: Find root causes. No temporary fixes. Senior developer standards.\n**Minimat Impact**: Changes should only touch what's necessary. Avoid introducing bugs.", "url": "https://wpnews.pro/news/boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents", "canonical_source": "https://gist.github.com/shahshrey/44da3ef7f0120297466c67bda526b018", "published_at": "2026-06-10 06:44:59+00:00", "updated_at": "2026-06-15 09:41:06.877179+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "large-language-models", "ai-research", "mlops"], "entities": ["Boris Cherny", "Claude Code", "CLAUDE.md"], "alternates": {"html": "https://wpnews.pro/news/boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents", "markdown": "https://wpnews.pro/news/boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents.md", "text": "https://wpnews.pro/news/boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents.txt", "jsonld": "https://wpnews.pro/news/boris-cherny-s-claude-code-workflow-as-a-drop-in-claude-md-workflow-subagents.jsonld"}}