By usedhonda
A collection of Claude Code skills for enhanced development workflows
Run skill evaluation against Golden Tasks
Generate agents and skills from conversation history
Initialize Golden Tasks for skill evaluation
Promote skill maturity level
Show current project knowledge status
並列タスク計画エージェント。プランモードで安全にコードベースを分析。 「Web並列」「web parallel」で発動。
タスク分解・並列作業計画の専門エージェント。 epicを分析し、並列実行可能なタスクカードを生成。 "タスク分解", "parallel plan", "epic分割"で発動。 <example> Context: ユーザーが大きな機能追加を依頼 user: "認証システムをOAuth2対応にして" assistant: "worktree-dispatcherでタスク分解します" <commentary> 複数の独立した作業に分割できる場合はこのエージェントで計画 </commentary> </example>
並列開発ワークフロー。 タスク分解→ブランチ作成→並列投入→auto-merge→レポート。 "/parallel-dev-orchestrator:par-plan", "並列開発", "タスク分解"で発動。
Analyzes Claude Code conversation history, git commits, and codebase to identify recurring patterns and generate project-specific agents and skills. Invoke with "/skill-from-history:learn-gen" or when user asks to "create skills from history", "analyze patterns", "generate agents".
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Learn project conventions from your conversation history, Git commits, and code. Generate reusable skills that encode your team's decisions—with evidence links.
# Add marketplace
/plugin marketplace add usedhonda/claude-skills
# Install
/plugin install skill-from-history@usedhonda-claude-skills
# Global (all projects)
mkdir -p ~/.claude/skills/skill-from-history
cp -r skills/skill-from-history/* ~/.claude/skills/skill-from-history/
# Project-specific
mkdir -p .claude/skills/skill-from-history
cp -r skills/skill-from-history/* .claude/skills/skill-from-history/
| Principle | Description |
|---|---|
| Project Learning | Learn from YOUR history, not generic best practices |
| Negative Learning | Learn what NOT to do from corrections |
| Evidence-first | Every pattern links to its source [E1][E2] |
| Context Engineering | Optimize for attention, not just tokens |
| Rules have half-life | Learning includes forgetting |
Conversations + Commits + Code
↓
Pattern Extraction
↓
Evidence Linking
↓
Human Review Gate
↓
Skill Generation
↓
Decay & Archival
"No, don't use Express. We use Hono now."
Your correction becomes a constraint. Next time, the AI already knows.
Tribal knowledge evaporates. claude-skills extracts it from actual behavior—with receipts.
Long contexts lose focus in the middle. We anchor critical decisions at the edges.
| Tool | Approach | claude-skills |
|---|---|---|
| Cursor Rules | Declare rules (human writes) | Observe patterns (history extracts) |
| Devin Knowledge | Store knowledge | Make knowledge executable |
| Generic linters | Enforce universal rules | Enforce YOUR conventions |
Project-Specific Skill Generator - Analyze your conversation history, Git commits, and codebase to generate project-specific agents and skills.
What it does:
Commands:
/skill-from-history:learn-status # Check current learning state
/skill-from-history:learn-init # Initialize Golden Tasks harness
/skill-from-history:learn-gen # Generate agents/skills from history
/skill-from-history:learn-check # Run evaluation against Golden Tasks
/skill-from-history:learn-promote # Promote validated constraints
Example workflow:
# 1. Check what patterns exist
/skill-from-history:learn-status
# 2. Generate skills from your history
/skill-from-history:learn-gen --skills
# 3. Output: .claude/skills/{name}/SKILL.md
Or say: "履歴からスキル作成", "analyze patterns", "learn from history"
Parallel Development Workflow - Decompose large tasks into parallel Claude Code sessions with auto-merge.
What it does:
Commands:
/parallel-dev-orchestrator:par-status # Show active plans and PR status
/parallel-dev-orchestrator:par-init # Check GitHub prerequisites
/parallel-dev-orchestrator:par-plan # Decompose epic → plan.yaml
/parallel-dev-orchestrator:par-dispatch # Submit tasks to sessions
/parallel-dev-orchestrator:par-harvest # Merge PRs and generate report
Example workflow:
# 1. Check environment is ready
/parallel-dev-orchestrator:par-init
# 2. Plan task decomposition
/parallel-dev-orchestrator:par-plan "Add OAuth2 authentication"
# 3. Dispatch to parallel sessions
/parallel-dev-orchestrator:par-dispatch
# 4. Watch and merge PRs
/parallel-dev-orchestrator:par-harvest --watch
Risk Policy:
| Level | Auto-merge | When |
|---|---|---|
| low | ✅ | Isolated changes, CI passing |
| medium | ✅ | + Required checks + scope isolation |
| high | ❌ | Auth, payments, data deletion → manual review |
Or say: "並列開発", "parallel tasks", "task decomposition"
Issues and PRs are welcome!
npx claudepluginhub usedhonda/claude-skills --plugin skill-from-historyComprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Complete developer toolkit for Claude Code
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.