Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
中文简介正在准备中。当前显示原文说明。
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
中文简介正在准备中。当前显示原文说明。
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
中文简介正在准备中。当前显示原文说明。
Use when you have a written implementation plan to execute in a separate session with review checkpoints
中文简介正在准备中。当前显示原文说明。
Use when executing implementation plans with independent tasks in the current session
中文简介正在准备中。当前显示原文说明。
Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".
中文简介正在准备中。当前显示原文说明。
Write ML papers for NeurIPS/ICML/ICLR: design→submit.
中文简介正在准备中。当前显示原文说明。
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
中文简介正在准备中。当前显示原文说明。
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
中文简介正在准备中。当前显示原文说明。
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
中文简介正在准备中。当前显示原文说明。
Karpathy's LLM Wiki: build/query interlinked markdown KB.
中文简介正在准备中。当前显示原文说明。
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
中文简介正在准备中。当前显示原文说明。
Comprehensive workflow for contributing features to open-source IDE and coding tool projects
中文简介正在准备中。当前显示原文说明。
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
中文简介正在准备中。当前显示原文说明。
Evaluate and install ecosystem tools (Hermes plugins, skills, integrations) — assess complexity vs. benefit before installation, show examples for visual tools, and avoid "cloned but not configured" states.
中文简介正在准备中。当前显示原文说明。
Search arXiv papers by keyword, author, category, or ID.
中文简介正在准备中。当前显示原文说明。
TDD: enforce RED-GREEN-REFACTOR, tests before code.
中文简介正在准备中。当前显示原文说明。
Write implementation plans: bite-sized tasks, paths, code.
中文简介正在准备中。当前显示原文说明。
4-phase root cause debugging: understand bugs before fixing.
中文简介正在准备中。当前显示原文说明。
Execute plans via delegate_task subagents (2-stage review).
中文简介正在准备中。当前显示原文说明。
Throwaway experiments to validate an idea before build.
中文简介正在准备中。当前显示原文说明。
Debug Python: pdb REPL + debugpy remote (DAP).
中文简介正在准备中。当前显示原文说明。
Pre-commit review: security scan, quality gates, auto-fix.
中文简介正在准备中。当前显示原文说明。
Plan mode: write markdown plan to .hermes/plans/, no exec.
中文简介正在准备中。当前显示原文说明。
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
中文简介正在准备中。当前显示原文说明。