an MCP (Model Context Protocol) server that gives AI coding agents the ability to see.
An MCP server that gives AI coding agents eyes. Turn any screen recording into a structured timeline your agent can reason over.
Install · Configure · Use · How it works · Roadmap
Coding agents like Claude Code, Codex, and Cursor reason well over text, code, and images — but they're blind to video. Developers share bugs, demos, and tutorials as screen recordings all the time. Watch bridges that gap: it analyzes a video and returns semantic events, not a frame-by-frame dump, so the agent can answer questions about what actually happened.
AI coding agents already have the brain. Watch gives them eyes.
A compact, structured timeline instead of thousands of tokens of frame descriptions:
cost reports real usage from the endpoint (via OpenRouter usage accounting); it's null for mode="manual" and getframes since those make no model call.
Event type is one of navigation, interaction, error, uichange, terminal, network, loading, dialog, codeedit, other.
watch(videopath, query=None, mode="sample", start=None, end=None) — analyze a video, return the timeline above plus a videoid. Modes:
mode what it does cost --- --- --- sample (default) AI scans scene-sampled frames → timeline cheap, flat manual no AI — just cache the video, return videoid + duration so the agent inspects frames itself zero full send the whole clip to a video-native model (start/end trims it first) scales with duration
full needs a video-capable endpoint. On OpenRouter that means e.g. qwen/qwen3.5-, qwen/qwen3.6-, or xiaomi/mimo-v2.5 — the qwen2.5-vl / qwen3-vl series are image-only and only work with sample
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/ManzarAli25--watch-it.html)
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is
zhuyansen/new-term-radarDaily radar for new AI terms: names of new models, products and repos, caught on X, Hugging Face, GitHub and sitemaps before keyword tools list them.
yzg1226/mcd-party-officer🍟 麦当劳门店派对成团调度 Skill — 把派对从「策划」推进到「办得成」：成团概率雷达 / 包场vs拼团 / 择日黄历 / 儿童营养指数 / 招募卡与邀请函
fellowgeek/mcp-memoryAn OKF-backed Model Context Protocol (MCP) server delivering persistent long-term memory and SQLite FTS5 search for AI agents.
hirotomasato/yowesGenerate realistic teacher documents (ID cards, licenses, letters) for 13 countries via MCP.
hujjun/mcd-saver-strategist麦当劳省钱点餐助手（麦门精算师）｜麦当劳优惠券组合优化 Skill：拉取麦当劳菜单与全部可用优惠券，算出最省的点餐组合，巨无霸套餐 ¥47 → ¥21.9（省 53%）。McDonalds MCP（麦当劳 MCP）coupon optimizer · McDonalds China MCP Server · Agent Skill for WorkBuddy · 麦当劳程序员创意开发大赛参赛作品
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.