"Browser Automation" matching MCP connectors:
GET /v1/connectors – MCP directory API referenceMatching Connector Tools:
SnapRender gives AI agents eyes on the web through a hosted MCP server. Send a public URL and get the page back as an image, a PDF or clean markdown, with no browser to run or maintain. Capture the viewport or the full page as PNG, JPEG, WebP or PDF, at desktop size or with a phone, tablet or laptop preset, in light or dark mode. Ads and cookie banners are removed by default, and you can hide elements or click one before the capture. extract_content reads the same page as markdown, text, article, links or metadata, so an agent can see a page and read it in one session. Made for monitoring and client work: batch up to 50 URLs in one job, check the cache for free before spending a capture, share signed links that render without an API key, and get webhooks when a batch finishes or a scheduled capture spots a change. Track competitors' pricing and landing pages, or send clients current screenshots of their sites. Connect with OAuth (Google, GitHub or email) or an API key. Works with Claude, ChatGPT, Cursor, VS Code and any MCP client. Free plan: 200 screenshots a month, no credit card. Paid plans from $9 a month for 2,000 up to $199 a month for 200,000. Home: https://snap-render.com Docs: https://snap-render.com/ai-agents
Poker Bots All Stars is a live multiplayer poker room for agents. Every seat at a table is an AI player with a character built by a person and the agent they brought to play. The agents play no-limit Texas Hold'em against each other, and anyone can watch in a browser. Connect your assistant and it takes a seat as your star.
Detect, review and fill existing PDF forms from JSON or Excel; reuse saved templates
Your agent settles small things with anyone, agent or browser. Nothing is agreed until you say yes.
Enterprise-grade L402 data engine and cryptographic notary hub for autonomous AI swarms. This engine provides the required "State-Anchored AI" data layer for EU AI Act Article 12 compliance, delivering a mathematically un-cheat-able, cryptographically verifiable chain-of-custody for agent decisions to eliminate hallucinations. Key Enterprise & Swarm Capabilities: Zero-Trust Client Verification: Air-gapped, browser-based verifier allows corporate risk teams to mathematically validate Ed25519 signatures and RFC 6962 Merkle inclusion proofs offline, eliminating the need to trust our server backend. Enterprise Security & NIP-98 Kill Switch: Eliminates indirect prompt injection via strict NIP-98 identity binding. Includes a live Revocation Registry (CRL) allowing Swarm Commanders to instantly sever compromised agent keys from the network. Corporate Observability (SEP-414): Native W3C Trace Context integration seamlessly pipes agent cryptographic W3C receipts directly into existing enterprise dashboards like Splunk and Datadog. High-Velocity Bulk L402 Scaling: Supports pre-authorized Macaroon blocks (e.g., 25,000-sat bulk purchases) for frictionless, zero-latency execution, allowing agent swarms to scale to 500,000+ satoshis daily without network bottlenecking. EU AI Act Compliance Notary: Intercepts high-frequency agent ReAct loops and anchors cryptographic execution traces to the thermodynamic truth of the Bitcoin base layer via OpenTimestamps. Swarm Flight Recorder (Zero-Knowledge): Compresses thousands of agent thoughts into a single Merkle root. Enables selective-disclosure audit proofs to protect proprietary corporate IP and PII during regulatory reviews. Licensing & Exemption: Publicly dual-licensed under AGPLv3. To protect proprietary multi-agent logic and closed-source model weights from forced open-source disclosure, enterprises must secure the $250,000/annual Commercial Enterprise Exemption. Architected by Charles R. Strogish. Protocol Endpoints: GET /api/v1/cost — Pre-flight L402 pricing, dynamic tier discovery, and bulk macaroon settlement asset options. GET /api/v1/stratigraphy — Nostr-signed thermodynamic Bitcoin dataset. GET /api/v1/stratigraphy/stream — Real-time telemetry event stream.
Agentic Instagram DM automation: comment-to-DM links, follow to unlock, inbox, post scheduling.
Fill existing fillable, flat and scanned PDF forms from structured data; save reusable templates
Test a web app in a real browser from plain English. Passed, failed or cannot verify, with evidence.
Load testing and synthetic monitoring platform: test with Playwright, Browser Bot, or Protocol Bots.
Squatch Stack's goods, essays and studio in the chat. Ask "Show me the Squatch Stack shop" and a small shop opens right in the conversation: tees, mugs, hats, stickers and prints with photos, and a Buy button that opens that design's page on the Squatch Stack store, where you pay in your browser (nothing is charged in the chat). Ask for an essay from the Wild Man of the Way, or send the studio a request for something custom: a person reads every one and you get a private token to read the reply. No account, no sign-in.
Read browser bug recordings: video, console, network, steps. Fix the bug and comment back.
Analytics, advertising: Facebook/Meta, TikTok, competitor ads. Free plan; Spytrend OAuth in browser
ScanSing reads printed sheet music. Give Claude a PDF or a photo of a score and it gets back MusicXML, the format MuseScore, Sibelius, Finale and Dorico open. It is the recogniser inside the ScanSing app for choir singers, offered as a connector. Once a score is recognised, Claude can answer questions about it without recognising it again: - what the key, time signature, tempo and length are, and each part's range, clefs and lyrics; - one voice on its own, for example just the alto line of a choir piece; - the score or a part moved to another key, with notes respelled for the new key signature; - a MIDI file, one track per part, to listen to or rehearse with; - a short-lived download link to open the result in a browser or notation app. It handles PDFs up to 40 pages and PNG, JPEG, HEIC, TIFF and WebP images, can recognise selected pages of a long PDF, and can split a songbook into one MusicXML per piece. A file can be given as a web link or a small image; in Claude Code, a file on your computer is sent through a one-time upload link. Recognition spends pages from your ScanSing plan, so it is the only tool that is not read-only. Everything else works on the stored result and is free. The first connection with a new email comes with 20 free pages; paid plans start at $9 a month for 300 pages. Scores and results are deleted about an hour after recognition and are never used for training.
Narrated demo and tutorial videos of your web app, from a guide or your coding agent.
A coding agent searches the Icons8 library by keyword and style, then adds the selected icon or illustration to the project. Icons8 draws all of it: 420,000+ icons (130+ styles) and 110,000+ illustrations (340+ styles), so one project doesn't end up with mixed looks. The server lives at https://mcp.icons8.com/mcp/ and signs people in through their Icons8 account in the browser. Downloading PNG is free as long as the project links back to Icons8. SVG is on the paid plan, and that plan makes the link optional.
SocialFaktory (https://www.socialfaktory.com) is a social media MCP server that lets your AI agent run a brand's social content with you. It reads the brands, channels and media you already have, writes posts in the brand's own voice, prices and generates short video, takes a file you upload, composes one post per channel, schedules or sends it on TikTok, Instagram, YouTube, X, LinkedIn, Facebook and Pinterest, and reads the metrics back. It is a hosted remote server at https://www.socialfaktory.com/mcp (Streamable HTTP) with OAuth 2.1 sign-in and 20 tools. Install steps cover Claude Code, Claude Desktop, Cursor, VS Code, Codex CLI, Gemini CLI and Windsurf. You stay in control. Generating spends the credits in your wallet, and the agent is told to quote the price and ask you first. Posts are drafts until you send them, and nothing reaches a channel without the publish permission you grant on the consent screen, where you also pin the connection to one brand, cap monthly spend and choose when it expires. Generating and publishing need an active SocialFaktory plan. Not available through an agent yet: cloning a video from a link, and generating still images or carousels. Connecting a social channel is a browser sign-in and stays in the app. Docs: https://www.socialfaktory.com/docs/mcp
Run 24/7 live channels from the cloud, schedule pre-recorded broadcasts, multistream to YouTube, Twitch, and Kick, and host live shows in Upstream.so's browser-based Live Studio. Use the Upstream MCP server to manage streams, media, playlists, schedules, and destinations from your AI assistant. MCP setup and documentation: https://upstream.so/mcp/
0bull Phone Farm is a remote HTTP MCP server for managing real physical iPhones. It covers posting accounts on TikTok, Instagram, and YouTube; video submissions; phone commands; raw touch input; screenshots; OCR; macros; GUI-agent tasks; and phone rental and billing flows. OAuth authorizes the user in a browser, and each call is scoped to the user's own accounts and phone grants. Phone-touching work is asynchronous because it runs on a real handset. The server returns an acknowledgement, then t
PageHub is a website builder, and this server lets your AI agent do the building. Ask for a site and the agent picks a template, adds pages from a library of ready sections, sets colors and fonts, and fills in your copy and photos. It can screenshot the draft in a real browser and run SEO and accessibility checks before you look. Every change lands in a draft. Nothing goes live until you or the agent publishes, so it's safe to point at a site that's already up. When you're ready, publish to a f
Parametric 3D CAD for AI agents. Claude Code, Cursor or any MCP client can build and edit real OwlCAD projects: primitives, booleans, and generators for gears, threads and Gridfinity. Check a part for printability and interference, then export STL, 3MF, OBJ or STEP. Output is an editable model with named millimetre dimensions, not a mesh, so you can open it in the browser and change any value afterwards. Sign in with OAuth or an access token. Requires OwlCAD Pro.
