7,463 mods in this category, of every kind an
agent can take. Each one carries what it costs per session, what the
scan found, and whether it is the original.
MCP server giving AI agents (Claude Code, Goose) control over Android devices via ADB — screen capture, UI automation, app launch, file transfer. Runs locally from the android-adb-mcp npm package.
★not rated 3▲
+1 1y agoA
tokens not measured
originalMIT
Control Zen for browsing, web testing, and debugging. Fill forms, capture network and console activity, take screenshots, and inspect console and network activity.
Screenshot any web page as an image (desktop or phone, cookie pop-ups blocked) or read it as markdown, with the SnapRender API. Use when the user wants to see, capture, compare or check how a website looks, or read a page.
Privacy-first, single-process web scraping for AI agents: MCP server plus a Firecrawl-compatible REST API, powered by local browser automation and your own LLM key. Runs locally from the firecrawl-lite-mcp-server npm package.
★not rated 3 2mo agoA
tokens not measured
originalMIT
ChangeDetection.io is a self-hosted or SaaS tool for monitoring web pages for changes — price drops, restock alerts, content updates, defacement monitoring. It's one of the most popular self-hosted tools (31k+ GitHub stars).
Fetch any web page's HTML or AI-parsed JSON, or run 100+ site plugins, via ScrapeUnblocker. Runs locally from the scrapeunblocker-mcp npm package. Needs 1 environment variable to run.
★not rated 3▲
+1 yesterdayA
tokens not measured
originalMIT
MCP server exposing the wavexis browser automation library to LLMs — 220 tools, 13 capability tiers, zero Node.js. Runs locally from the wavexis-mcp Python package.
★not rated 3 4d agoA
tokens not measured
originalMIT
Plain-English browser tests for AI agents: real Chrome, deterministic verdicts. Free, no API keys. Runs locally from the browserbash-cli npm package. Needs 1 environment variable to run.
Top-level Windows computer-use skill with a bundled standalone runtime that bootstraps itself without any local Claude installation, private native modules, or extracted app assets.
Playwright-style MCP server for pywebview apps — lets Claude see, control, and debug your Python desktop web UI without modifying your app's source code.
Manual QA runs driven through a browser — derive the scenarios from the ticket and the diff or execute the list you were handed, write them down as a scenario file under docs/qa/ / with the test data they need, preconditions and dependencies, execute them with Playwright in a QA window that stays open between runs.
Connects Claude Code to ToolPiper, the on-device AI platform for macOS: local LLM inference, speech to text, vision and OCR, browser automation, video editing, and system control through one loopback MCP server.
★not rated 3
changed 21d agoA
tokens not measured
originalMIT
Record a narrated demo video of a web flow by driving agent-browser. Use when asked for a product walkthrough, feature demo, tutorial video, or "record a video of X happening in the browser". Output is an mp4 with per-step narration.
A job-search skill that searches current public listings using an existing résumé and job requirements, checks sources, removes duplicates, and compares roles.
Web scraping, search, crawling, and structured extraction for Cursor agents through MCP.
★not rated 3 1mo agoA
tokens not measured
AGPL-3.0
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: