도구

Tencent's open-source browser bridge: through the bsk CLI and a Chrome/Edge extension, agents use your logged-in accounts in a separate visible window to browse, fill forms and take full-page screenshots, and correlate network requests with console output to debug sites.

Open-source research agent: one command runs literature reviews, deep research, paper reviews, paper-code consistency checks and replication plans, searching arXiv, PubMed and more with sources recorded. Good for grad students writing reviews. Run feynman setup to sign in to a model provider.

Anthropic's official AI governance plugin: tiers AI use cases against your internal register, runs impact assessments across applicable laws such as the EU AI Act, reviews vendor AI terms for training-data and liability risk, and periodically checks for drift between AI policy and practice.

NVIDIA's open-source LLM vulnerability scanner that probes for prompt injection, jailbreaks, data leakage, hallucination and harmful content, supporting OpenAI, Hugging Face, Bedrock and most REST-accessible models. Scanning cloud models needs their API keys.

Privacy compliance plugin: decides whether processing activities need a PIA or DPIA, generates privacy impact assessments, reviews DPAs from the controller or processor side, and drafts DSAR responses within statutory deadlines. Official from Anthropic.

Annotate and review agent-generated plans, docs and code diffs in the browser, then send feedback back to the agent in one click. Hooks into plan mode and supports Claude Code, Codex and other agents. Run the official install script to set up the CLI first.

Extracts the main content of a web page, stripping comments, sidebars, headers and footers, and outputs clean HTML or Markdown with metadata like title, author and publish date, saving tokens when agents read the web. Built for Obsidian Web Clipper by kepano; still in development.

Official AWS MCP that searches and reads AWS documentation, converts it to Markdown for agents to cite, and suggests related pages. No AWS credentials needed; requires uv.

Rendering engine that lays out HTML and CSS and exports PDF, supporting paged media and other print style standards, ideal for generating reports, invoices and receipts in bulk. Written in Python; depends on the system Pango library.

High-accuracy retrieval-augmented Q&A over paper PDFs, text and Office documents, filling in citation metadata, checking retractions and attaching traceable sources to answers. Provides the pqa CLI and a Python library. Needs an LLM API key by default; local models can be configured.

Uses a strong model to audit your codebase and list evidence-backed issues across correctness, security, performance and testing, writes a plan a cheaper model can execute, and reviews the results. It never edits code itself. By shadcn.

Say build a harness for this project and it designs a domain-specific agent team, generating subagent definitions, companion skills and orchestration, validates them and keeps evolving them from feedback. Supports workflow orchestration, persistent agent teams and subagent delegation.

Static scanner for security and compliance issues in Terraform, CloudFormation, Kubernetes, Helm, Dockerfiles and CI pipeline configs, with over a thousand built-in policies and hard-coded secret detection. Outputs SARIF, JSON and more. Maintained by Prisma Cloud.

Manages multiple Claude Code, Codex, Gemini and Aider sessions in parallel from one terminal: each task gets its own git worktree, can run in the background, and changes are reviewed before merging. Requires tmux and gh.

Companion skills for Google Stitch: generate UIs from text or images, extract DESIGN.md from code, and turn designs into React or React Native components. From Google Labs but not an officially supported Google product; set up the Stitch MCP first.

Free remote MCP service that turns any GitHub repo or GitHub Pages site into a docs endpoint, so agents can read up-to-date docs and search code on demand and hallucinate fewer APIs. No sign-up or install; self-hosting is also possible.

Spec-driven project management for software: guides brainstorming into a PRD, breaks it into epics and dependent tasks synced as GitHub Issues, answers standup, next-step and blocker questions, and can dispatch several agents to code in parallel. Needs Git, an authenticated gh CLI and a GitHub repo.

Offline speech toolkit for Apple Silicon Macs: CLI text-to-speech (including Chinese and voice cloning) and speech-to-text, for voicing lessons and explainer videos or transcribing lectures and interviews, with data staying local. Needs an M-series Mac and Python 3.10+; MP3 export needs ffmpeg.

Turns Peng Sida's public research-writing notes into a skill: revises abstracts, introductions, methods and experiments of ML/CV/NLP papers from a reviewer's view, checks that each claim has experimental evidence, and runs a pre-submission self-check. Pure methodology, no API key.

HTTP client CLI that reimplements HTTPie's syntax in Rust with a focus on performance; key-value pairs on the command line are sent as JSON by default, making it quick to debug REST APIs.

Moonshot AI's official terminal coding agent: reads and edits code, runs commands, searches files and fetches web pages, with video input, MCP, plugins, subagents and hooks, plus Zed and JetBrains integration via ACP. Needs a Kimi sign-in or Moonshot API key. Replaces the archived Kimi CLI.

Removes backgrounds from images and videos with one AI command: batch whole folders, use a portrait model, refine edges or swap in a solid color, and output transparent MOV or composite video onto a new background. Runs locally; install PyTorch and ffmpeg first. Models download on first run.

Trail of Bits static analysis skills: CodeQL for interprocedural taint tracking and data-flow analysis, Semgrep for rule-based scanning, plus parsing and deduplicating SARIF results. Requires CodeQL or Semgrep installed locally.

Trail of Bits skill that audits GitHub Actions workflows wired to AI coding agents (Claude Code Action, Codex, Gemini CLI and others) for prompt injection, env-variable relays, dangerous sandbox settings, wildcard allowlists and other privilege risks.