发现 Skills

Build UIs with shadcn/ui components. Use when adding accessible, styled React components, building a design system with Tailwind, customizing component libraries, or scaffolding UI for Next.js apps.

HashiCorp Vault for secrets management and data protection. Use when the user needs to store and access secrets, manage encryption keys, configure dynamic credentials, handle PKI certificates, or integrate secrets into applications.

Verify an OptimizeSpec self-improvement implementation. Use when checking generated skills, artifact completeness, eval runner behavior, ASI quality, direct eval, compare, or GEPA optimize readiness.

Template for the required sections (Inputs / Outputs / Forbidden / Verification / Evidence) when creating a new Skill. Referenced by the skill-builder when generating a new Skill.

Cut a Han release: update CHANGELOG.md with the changes since the last release, tag the version vX.Y.Z, and publish a GitHub release whose notes attribute every merged pull request to its author and link back to the full changelog for that version. Use when releasing, cutting a release, shipping a new Han version, publishing release notes, or tagging a version. Reads the target version from plugin.json; when the version has not been bumped past the latest tag yet, it proposes a semantic-versioning bump and confirms it before continuing. Requires the gh CLI, jq, and a clean git checkout. This is a repository-maintenance skill for the Han repo itself, not a general review or PR skill — use code-review for local review, gh-pr-review to post a PR review, and update-pr-description for PR bodies.

Discovers key attributes of the current code repository and its projects — languages, frameworks, tooling, configuration, documentation structure — and writes a static reference for other skills, agents, and hooks to consume. Use when scanning, analyzing, or detecting the project's technology stack, build tools, or repository structure. Does not create or update project documentation — use project-documentation for writing feature or system docs.

汇总当日（或指定日期）的 AI 助手对话记录，生成格式化工作日志， 直接在对话框中输出。 触发词：每日汇报、汇报一下、昨天聊了什么、昨天干了啥、昨天总结、 今日汇报、今日总结、每日总结 参数：可附加日期，如"每日汇报 04-21"，省略则默认当天 输出：B模式（详细），包含所有会话主题、关键操作、问题描述、状态标注。

Launch Work - task management and personal assistant (PA) support for work

Check mission status and progress via a fresh subagent. Queries Beads for features, tasks, bugs, and recent work activity. Keeps HOUSTON context lean.

Expert patterns for Godot 3D lighting including DirectionalLight3D shadow cascades, OmniLight3D attenuation, SpotLight3D projectors, VoxelGI vs SDFGI, and LightmapGI baking. Use when implementing realistic 3D lighting, shadow optimization, global illumination, or light probes. Trigger keywords: DirectionalLight3D, OmniLight3D, SpotLight3D, shadow_enabled, directional_shadow_mode, directional_shadow_split, omni_range, omni_attenuation, spot_range, spot_angle, VoxelGI, SDFGI, LightmapGI, ReflectionProbe, Environment, WorldEnvironment.

Expert patterns for AnimationPlayer including track types (Value, Method, Audio, Bezier), root motion extraction, animation callbacks, procedural animation generation, call mode optimization, and RESET tracks. Use for timeline-based animations, cutscenes, or UI transitions. Trigger keywords: AnimationPlayer, Animation, track_insert_key, root_motion, animation_finished, RESET_track, call_mode, animation_set_next, queue, blend_times.

Pre-trade risk gates and post-trade audit actions for institutional compliance.

Foundation layer for identity verification, noise filtering, and environment awareness.

Consult Bun's official documentation snapshot for runtime APIs, `bun` CLI commands, `bunfig.toml`, package management, bundling, testing, deployment guides, and ecosystem integrations. Use this whenever the user asks how to do something with Bun, troubleshoot Bun-specific behavior, compare Bun to Node/npm, or wants Bun command or API guidance, even if they only mention a command like `bun install`, `bun test`, `bun build`, `bunx`, `Bun.serve`, or a Bun config option.

Discover, document, and map all API endpoints of an Android application

Use this skill when orchestrating multi-agent swarms with Claude Code's TeammateTool and Task system. It applies when coordinating multiple agents, running parallel code reviews, creating dependency-aware pipeline workflows, building self-organizing task queues, or any task that benefits from divide-and-conquer patterns.

Control Home Assistant - smart plugs, lights, scenes, automations.

Write and debug AST-based codemods using jscodeshift for automated code transformations. Use when creating migrations, API upgrades, pattern standardization, or large-scale refactoring.

Markdown-based presentation slides powered by Vue and Vite. Use when creating slide decks, technical presentations, conference talks, or any presentation that benefits from code highlighting, diagrams, and web technologies.

Project-specific development instructions. Use whenever instructions are mentioned. Auto-load when working on tasks like coding, testing, git operations, or exploring the codebase. Read the matching file from references/.

Install Python Clean Architecture development agents into a local project. Fetches latest agent definitions from GitHub and installs to .claude/agents/. Use when setting up architect, domain-modeler, implementer, test-engineer, and code-reviewer agents in a new or existing project.

Use when generating or updating user story YAML files for UI testing — browses a site, discovers flows, creates structured stories with diff-aware updates, negative testing, and self-validation. Keywords - stories, generate, create, user journey, persona, QA, testing.

Build and modify apps in the Windows 15 desktop environment. Use when creating new apps, modifying existing apps in apps/, working with window management, persistence, hotkeys, notifications, or UI components.

Execute and troubleshoot the Supabase edge function `process_hybrid_search`, which rewrites process descriptions and calls `hybrid_search_processes` with optional filters. Use when validating process query/filter behavior, tuning retrieval prompts, or debugging auth, embedding, and RPC failures.