rohitg00/pro-workflow

这个仓库里有 41 个技能，GitHub 星标 ★ 2,870。

• agent-teamsCoordinate multiple Claude Code sessions as a team — lead + teammates with share
• auto-setupAuto-configure quality gates, hooks, and settings for a new project. Detects pro
• batch-orchestrationDecompose large-scale changes into independent units and spawn parallel agents i
• bug-captureCapture a user-reported defect as a durable GitHub issue written in the project'
• compact-guardSmart context compaction with state preservation. Saves critical files, task pro
• context-engineeringMaster the four operations of context engineering — Write, Select, Compress, Iso
• context-optimizerOptimize token usage and context management. Use when sessions feel slow, contex
• cost-trackerTrack session costs, set budget alerts, and optimize token spend. Use to check c
• design-engineeringApply interface craft when building or reviewing UI - motion, easing, timing, sp
• deslopRemove AI-generated code slop, unnecessary comments, and over-engineering from t
• domain-modelingBuild the project's shared language and bounded contexts before writing code, so
• file-watcherConfigure file watching hooks to auto-react to config changes, env file updates,
• improve-architectureAudit an area of the codebase and propose the smallest structural moves that imp
• insightsShow session analytics, learning patterns, correction trends, heatmaps, and prod
• learn-ruleCapture a correction or lesson as a persistent learning rule with category, mist
• llm-councilProvider-agnostic multi-LLM deliberation. Three phases — independent responses,
• llm-gateLLM-powered quality verification using prompt hooks. Validates commit messages,
• mcp-auditAudit connected MCP servers for token overhead, redundancy, and security. Use wh
• module-mapProduce a one-screen map of an unfamiliar area of the codebase: entry points, mo
• orchestrateWire Commands, Agents, and Skills together for complex features. Use when buildi
• parallel-worktreesCreate and manage git worktrees for parallel coding sessions with zero dead time
• permission-tunerAnalyze permission denial patterns and generate optimized alwaysAllow and always
• plan-interrogateStress-test a plan by walking its decision tree one question at a time. Use when
• pro-workflowComplete AI coding workflow system. Orchestration patterns, 18 hook events, 8 ag
• replay-learningsSurface past learnings relevant to the current task before starting work. Search
• safe-modePrevent destructive operations using Claude Code hooks. Three modes — cautious (
• session-handoffGenerate a structured handoff document capturing current progress, open tasks, k
• skill-optimizerSkillOpt-flavored offline training loop for any SKILL.md. Treats accumulated lea
• skill-routerThe index of every pro-workflow skill and command, grouped by job, with when to
• smart-commitRun quality gates, review staged changes for issues, and create a well-crafted c
• sprint-statusTrack parallel work sessions and prevent confusion across multiple Claude Code i
• survey-generatorCompile a structured literature survey on any AI/ML topic. Agent curates a resea
• tddDrive a change through a red-green-refactor loop - failing test first, minimal c
• thoroughness-scoringScore every decision point with a Thoroughness Rating (1-10). AI makes the margi
• token-efficiencyReduce token waste by 40-60% through anti-sycophancy rules, tool-call budgets, o
• wiki-builderStart, structure, and grow a persistent research wiki indexed in pro-workflow's
• wiki-queryQuery pro-workflow wikis via SQLite FTS5 BM25 retrieval. Returns top-K passages
• wiki-research-loopAuto-grow a pro-workflow wiki by running a budget-capped BFS research loop over
• wiki-viewerRender a self-contained HTML viewer for a pro-workflow wiki. Pages, sources, cla
• wrap-upEnd-of-session ritual that audits changes, runs quality checks, captures learnin
• writing-guidelinesApply clear-writing standards to any prose the agent produces - READMEs, docs, U