🔌
Plugins
Loading...
Showing 30 of 22991 results
Skillを自動的に成長させるためのプラグイン。セッション中のユーザー発言・承認コマンド・スキル発火をセッション別ディレクトリに記録し、セッション終了時にやり取りの要約も書き出す。蓄積されたログから新しいスキルを作成する3つの分析スキル（複数セッション横断のパターン検出、単一セッションからのスキルdraft生成、承認コマンド分析）を提供する。
Local, model-agnostic automatic skill injection: embeds the prompt locally, ranks it against skill descriptions, and injects the matched skill via UserPromptSubmit/PostToolUse/SessionStart hooks (ski). Requires the `ski` binary — cargo install --path .
Design, evaluate, and improve Agent Skills with intent compilation, isolated evaluation, and deterministic validation
Skill 创建预检与可靠性验收工具，支持旧版指令失稳识别、硬要求来源发现、Ed25519-signed 外部基线与 held-out、逐约束追踪、验证模态与产物阶段匹配、候选/producer 绑定运行证据、最终回执验签、多轮漂移门禁、七层 Harness、业务流和安全风险审查
Maintenance tools for skills repos: validation, quality, freshness, upstream detection, best practices review, wiki-sanity lint
Skill management utilities — load, install, and register skills into the local plugin system
Manage your Claude Code skills and plugins from inside any project — switch which plugins are active, publish and sync local skills to a central marketplace, and diagnose and auto-repair a broken setup. Adapts to any repo layout and delegates skill authoring to the native skill-creator.
Three-tier activation manager for agent assets (skills, subagents, commands, workflows): keep a large library dormant at zero context cost and activate only what each project needs, via a symlink/junction/tracked-copy ladder with drift detection and repair.
管理 Claude Code、Codex 和 OpenClaw Skills/Commands 的安装、同步、卸载和列表查看
Personal Agent Skills collection for Claude Code and Codex.
Package AI agent skills into deployment formats — .zip, .skill, Claude plugin, Claude marketplace, Cursor plugin, Cursor marketplace, ChatGPT/Manus zip, Codex CLI, NanoClaw marketplace, OpenClaw/ClawHub, Agent Skills standard (.agents/), or a universal repo with all formats + CI/CD.
Dynamic skill quality evaluator — fork target skills, enforce JSON output via --json-schema, run N-repeat against ground-truth snapshots with semantic-judge + calibration, surface regressions vs baseline + auto-propose SKILL.md edits
Iterative skill review loop using 4 specialized parallel review agents. Reviews discoverability, structure, clarity, and best practices, then fixes and repeats until clean.
Review and optimize Claude Code skill files — score SKILL.md quality, plan and scaffold new skills, audit allowed-tools permissions, optimize frontmatter (descriptions + disable-model-invocation), and enforce the three-part 200-char description format
Review and optimize Claude Code skill files — score SKILL.md quality, plan and scaffold new skills, audit allowed-tools permissions, optimize frontmatter (descriptions + disable-model-invocation), and enforce the three-part 200-char description format
Security audit and vulnerability scanner for AI agent skills. Scans for malicious code, prompt injection, data exfiltration, supply chain risks, and privilege escalation before installation. Zero dependencies, PASS/WARN/FAIL verdicts with remediation guidance.
Comprehensive productivity skills and development tools for Claude Code, including git workflow automation, code context understanding, peer LLM consulting, and PR review feedback processing.
Keep locally installed Claude Code skills in sync with their GitHub repos
1 command + 1 skill (/skill-test). Orchestrates a 5+1 stage pipeline (Triage → static review → behavioral eval → regression → looper) for skills/agents, and a 7+1 stage pipeline for patterns. Supports --from-stage for resuming mid-pipeline.
Auto-syncs docs/skills.md entries to a global cross-project skills log with a local search viewer on port 38888.
Reusable Claude Code skills: handoff write/resume, UI-to-functional planning, tiered brainstorm→build, native+web UI/UX design, and agent-driven branch merge/prep