Tools

Lets Claude Code push ML research forward unattended: literature review, topic selection, experiment deployment, cross-model review loops and paper writing through rebuttal, in 80+ skills. Codex acts as the independent reviewer by default; other model pairings work too.

Browser control from the browser-use team: an editable CDP connection lets agents drive the Chrome you are already using, writing missing helper functions as they go and keeping them for reuse. Also offers an MCP server. Allow remote debugging in Chrome on first use.

Context and harness engineering skills: context fundamentals, degradation diagnosis and compression, multi-agent architectures, memory systems, tool design, LLM-as-judge evaluation and self-improvement loops. Focused on principles and design patterns, not tied to any agent platform.

Context and harness engineering skills covering context compression and optimization, multi-agent architectures, memory systems, tool design and LLM-as-judge evaluation, activated automatically by task. Aimed at building production-grade agent systems.

Official Lark (Feishu) skill to create and edit Lark Sheets: formulas, charts and pivot tables, conditional formatting and filters, plus import and export of local Excel/CSV, for summaries, budgets and reports. Needs lark-cli on a machine with Node.js, plus app setup and authorization.

Official Lark (Feishu) approval skill: list pending, handled and submitted approvals; approve, reject, transfer, return, remind or withdraw; and find and submit native approvals such as leave or expense requests. Needs lark-cli with app setup and sign-in authorization.

Write end-to-end UI tests for Android, iOS and web apps in YAML. The CLI includes an MCP server (maestro mcp) so coding agents can tap, assert and generate tests on simulators. Requires Java 17+.

MCP built into the CLI of the Maestro mobile UI testing framework (maestro mcp): lets agents view the screen, tap, scroll and assert on iOS/Android simulators, Android devices or Chromium, and write and run YAML end-to-end flows. Needs the Maestro CLI and Java 17+. Official.

Simplifies Figma designs into layout and style data for coding agents, so UIs built from the design are more accurate than working from screenshots. Requires a Figma personal access token.

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.

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.

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.

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+.

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.

MiniMax's official Excel skill: read, analyze, create, edit and validate xlsx files and fix broken formulas. Edits existing files at the XML level to keep formatting, formulas and pivot tables; writes results as formulas with finance-style color coding. Needs Python (pandas); LibreOffice optional.

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.

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.

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.

Lets agents operate the Windows desktop directly: launch apps, switch windows, simulate keyboard and mouse, read UI elements and run PowerShell, without needing a vision model. Windows only.