Tools

Version datasets and models the way Git versions code: data lives in remote storage like S3 or GCS while versions are tracked in your Git repo. Also offers reproducible pipelines and local experiment tracking, rerunning only the steps affected by a change.

ModelScope's official training framework for LLMs and multimodal models: the swift command covers pretraining, fine-tuning, RL, inference, evaluation, quantization and deployment for 600+ text and 400+ multimodal models, with Megatron parallelism and a web UI. Usually needs a GPU.

Multi-language git hook manager: declare formatting, lint, secret scanning and other hooks in .pre-commit-config.yaml to run automatically before each commit, or have the agent self-check with pre-commit run --all-files before committing.

Loads and unloads environment variables per directory: reads .envrc (and optionally .env) when you enter a project and restores the environment when you leave, so configs and secrets do not leak between projects. Hook it into bash, zsh, fish or another shell first.

Open-source zero-shot TTS via f5-tts_infer-cli: from a few seconds of reference audio it reads Chinese or English text in the same voice, handy for lessons and videos. Needs Python, PyTorch and FFmpeg, ideally a GPU. Pretrained models are non-commercial; get consent before cloning a voice.

Turns 18 kinds of sources, including doc sites, GitHub repos, PDFs and videos, into skill packages for Claude and other platforms, or exports RAG chunks and vector store data. Optional MCP server, and it can call a local agent like Claude Code to enrich content. Needs Python 3.10+.

Audits a whole React codebase read-only: runs a React Doctor scan as evidence, vets findings across bugs, performance, accessibility, security and maintainability, ranks them by leverage, then writes self-contained plans other agents can execute. Never edits source.

Runs React Doctor via npx to scan React code for security, performance, correctness, accessibility and architecture issues with a 0-100 health score; checks changed code for regressions, and on /doctor fetches the official triage playbook to fix issues.

Writes precise prompts for a specific AI tool: identifies the target tool, extracts your intent, asks at most three questions if details are missing, then outputs a lean, paste-ready prompt. Covers dozens of tools including ChatGPT, Claude Code, Cursor, Midjourney and Sora.

Companion skill for the official Playwright CLI: short commands to open pages, click and fill forms, take snapshots and screenshots, and write and debug Playwright tests, using less context than the MCP approach. Maintained by Microsoft; needs Node.js.

Microsoft Playwright's official browser CLI for coding agents: open pages, click and type, take screenshots and manage sessions, and install the companion skills in one step. Uses fewer tokens than Playwright MCP. Needs Node.js 18+.

Nearly 100 skills spanning the AI research workflow: ideation and literature review, fine-tuning and post-training, distributed training, inference serving, evaluation, interpretability and paper writing, tied together by an autoresearch orchestration skill.

Plugin marketplace for AI research with nearly 100 skills in 23 categories, from ideation, fine-tuning and post-training to distributed training, inference serving, evaluation and paper writing, coordinated end to end by the autoresearch skill.

MiniMax's official skill plugin with dev guides for frontend, full-stack, Android, iOS, Flutter, React Native and GLSL shaders, plus PDF, PowerPoint, Excel and Word document skills and image, video and music generation. Media skills need a MiniMax API key. Still in beta.

MiniMax's official skill collection covering frontend, full-stack, Android, iOS, Flutter, React Native and GLSL shader development, plus PDF, Word, Excel and PowerPoint document generation. Some media skills need a MiniMax API key. Still in beta.

Customizes the Claude Code status line to show model, Git branch, token and context usage, session and weekly limits at the bottom of the terminal. Configured through an interactive TUI with Powerline styling. Claude Code only.

Runs an extremely strict maintainability review of the current branch: hunts for restructurings that make the change dramatically simpler and flags files pushed past 1,000 lines, ad-hoc conditionals, thin wrappers and needless casts. From Cursor's official plugins; invoke manually.

Simon Willison's LLM CLI and Python library: call OpenAI, Claude, Gemini and local models from the terminal, with conversations logged to SQLite automatically, plus embeddings, structured extraction, tool calling and a plugin system.

From Zilliz: indexes your codebase as vectors in Milvus or Zilliz Cloud so coding agents find relevant code by semantic search, cutting context costs in large repos. Needs an embedding model (OpenAI, VoyageAI, Ollama and others) and a vector database.

A du replacement written in Rust that shows disk usage as a directory tree with proportion bars, making large files and folders easy to spot. Filter by depth, count or regex, and output JSON with -j for agents to parse.

Lets agents call thousands of ready-made scrapers and automation Actors from Apify Store to pull data from social media, search engines, maps, e-commerce sites and more. Maintained by Apify; needs an Apify account (OAuth) or API token.

ByteDance's open-source, general software engineering agent: trae-cli edits files and runs commands from natural language, supports multiple model providers and MCP, and records full execution trajectories. Transparent design suits agent research and customization. Needs your own model API key.

Bundles 60+ full-stack specialist skills into one plugin: experts for 12 languages including Python, TypeScript, Go, Rust and Java, and frameworks like React, Django and Spring Boot, activated automatically by request. Also includes Jira and Confluence workflow commands (needs Atlassian MCP).

Connects to the Arize Phoenix observability platform so agents can query LLM traces and sessions, manage prompts, browse datasets and read experiment evaluation results. Maintained by Arize; needs a self-hosted or cloud Phoenix instance.