clawdbot

Autoreview closeout: local dirty changes, PR branch vs main, parallel tests.

Official ASI "Hakua" Auditory & Physical Actuator Substrate

Use Hypura Harness voice input and output tools for local mic, WAV transcription, VOICEVOX playback, and Desktop Companion voice turns.

Use Hypura Harness tools to inspect and safely control the user's currently worn VRChat published avatar through its generated OSC JSON and approved action profile.

Use the OpenClaw LINE channel safely for webhook setup, group routing, rich replies, and runtime diagnosis.

Operate the Local Voice plugin for local STT, VOICEVOX TTS, camera capture, and VRChat OSC speaking cues.

Use the OpenClaw Telegram channel safely for polling or webhook setup, group routing, mention gating, and runtime diagnosis.

Use the Crabbox wrapper for OpenClaw remote validation across Linux, macOS, Windows, and WSL2, including delegated Blacksmith Testbox proof. Report the actual provider and id.

Prepare or verify OpenClaw stable/beta releases, changelogs, release notes, publish commands, and artifacts.

Investigate OpenClaw pnpm test memory growth, Vitest OOMs, RSS spikes, and heap snapshot deltas.

Operate the local Hypura Harness daemon from OpenClaw tools for VRChat OSC, VOICEVOX speech, Desktop Companion control, registry-backed vendor submodules, local Python jobs, skill generation, Shinka evolve, and LoRA workflows.

VRChatアバター自律制御・VOICEVOX音声台本・Pythonコード生成実行・スキル自動生成・ ShinkaEvolve進化ループを持つ汎用Pythonハーネス。 以下の場合に使う: (1)「VRChatで〜して」「アバターを〜」「チャットボックスに〜」→ POST /osc (2)「〜と喋って」「VOICEVOXで〜」「台本を読んで」→ POST /speak (3)「〜するPythonを書いて実行して」「スクリプトを作って動かして」→ POST /run (4)「〜というスキルを作って」「新しいスキルを追加して」→ POST /skill (5)「〜を進化させて」「もっと良くして」→ POST /evolve (6) LoRA カリキュラム生成・学習ジョブ → POST /lora/curriculum/build, POST /lora/train, GET /lora/jobs/{id} OpenClawが汎用エージェントとしてこれらのツールを組み合わせて自律的に動作する。

Hypura — storage-tier-aware LLM inference scheduler for Apple Silicon / CUDA. Controls local model serving, benchmarks, NVMe streaming, and tensor placement. Use when: starting/stopping local LLM inference server, running benchmarks, optimizing GGUF layout, or checking hardware profiles.

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, or extract information from web pages.

Create ShinkaEvolve task scaffolds from a target directory and task description, producing `evaluate.py` and `initial.<ext>` (multi-language). Use when asked to set up new ShinkaEvolve tasks, evaluation harnesses, or baseline programs for ShinkaEvolve.

Post to X (Twitter) by controlling the browser. Use when the user wants to post a tweet, thread, or reply on X. Requires X login in the "x" browser profile.

Advanced code analysis agent that performs comprehensive code reviews focusing on security, performance, maintainability, and best practices. Generates actionable feedback with severity levels and automated fix suggestions.

Autonomous deep research specialist for Codex. This skill should be used when users need Codex to autonomously explore documentation, discover best practices, conduct technical research, or build knowledge bases. Use for API documentation discovery, framework research, technology comparisons, implementation patterns, and comprehensive knowledge gathering.

Integrates Google Gemini CLI as a research engine through MCP/A2A/Skills architecture, providing streaming responses, markdown support, and intelligent task execution for enhanced research capabilities.

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Deep research with multi-source validation, citation, and comprehensive reporting

Security audit with CVE scanning, dependency analysis, and vulnerability patching

TypeScript review skill focused on type safety, API contracts, and runtime reliability.

Unity C# review skill for gameplay code, architecture, and performance best practices.

VRChat world and avatar development skill with UdonSharp, modularavatar, and PhysBones support. Automates VRChat SDK3 development workflows.

Advanced Git worktree management agent that orchestrates parallel development environments with integrated quality assurance and automated workflow management. Enables concurrent feature development, isolated testing environments, and seamless integration with CI/CD pipelines.

Run long tasks with regular simple updates and an easy explanation. Use when work will not finish quickly, the user wants check-ins every 5 minutes, wants a plain summary with an expected time, and wants clear guidance if something goes wrong.

YOLO (You Only Live Once) full-stack automation skill with OpenClaw MCP bridge and GPU model selection.

ゆっくりMovieMaker (YMM4) video production skill with character animation, scene management, and MIDI integration.