Watch Skill plugin for Grok Build
Give AI agents eyes, ears, and verifiable results. Watch Skill turns video, audio and screen activity into searchable, timestamped evidence and proves work with deterministic contracts, not model opinion. DeepWatch is the agent workspace built on DeepSeek Harness. Python + npm, MCP, CLI, REST, Web.
How to install Watch Skill in Grok Build
grok plugin install oxbshw/watch-skill --trustReview before you trust it. --trust lets this plugin's hooks, MCP servers and scripts run code on your computer. It's a community repo that neither xAI nor GrokStacks has reviewed, so read its code first.
We install this repo directly: Grok Build 1.0.62 can't install it through its own marketplace file. Then start a new Grok session, or press r in /plugins, to load it.
Command format: Grok Build's plugin guide
Runs a local MCP server (with npx, uvx or similar), so that tool needs to be installed, for example Node.js for npx.
What we checked: on Oct 18, 2026 it installed with grok 1.0.62 and Grok found 11 skills, 1 slash command. That's an install check only: we haven't run it in a live Grok session.
What it contains
Skills (11)
asking-with-evidence— The user asks a question about a video that was already watched or indexed — "what did they say about X", "what error code appears", "what happens at 2:30", "does the video show Y". Use this to answer from the persistent index with timestamconfiguring-vision— The user wants to connect an LLM or vision provider, already has an API key, asks "can I use OpenAI/Anthropic/Gemini/OpenRouter", wants local Ollama, or needs different cheap and strong models. Use this to configure provider-neutral visualextracting-structure— The user wants structure pulled out of a watched video — "make chapters for this video", "where does the bug appear in this recording", "turn this screen recording into a bug report", "how strong is my intro/hook". Use this for deterministilearning-from-mistakes— The user corrected an answer about a video — "no, it actually says X", "that's the wrong timestamp", "you misread the error code" — or asks why a video answer was wrong. Use this to record the correction as a lesson so future answers on simrecovering-from-errors— A Watch Skill command failed, video download broke, ffmpeg or yt-dlp is missing, a provider returned an error, local vision stopped, or the user asks "why can't it watch this" or "repair my setup". Use this to diagnose and apply the engine'sharing-results— The user wants to share or hand off what was found in a video — "send this analysis to my team", "make a page I can share", "export the findings", "give me something I can attach to the ticket". Use this to render a self-contained offline Hthe-loop— The user built or changed something visual — a UI, an animation, a game, a generated video — and wants it verified, or asks "why does my UI look wrong", "check that the fix actually worked", "does the animation glitch". Use this to record tvideo-memory— The user asks about videos watched in the past or across sessions — "have we watched anything about X", "which video showed that error", "what did that meeting decide", "search my videos", or a question that spans several videos. Use this twatch— Watch any video (URL, stream, or local path) via Watch Skill. Downloads, extracts scene-aware deduped frames, OCRs them, transcribes (captions first, then local Whisper — offline by default), indexes everything, and hands the result to thewatching-videos— The user shared a video URL, a YouTube/TikTok/stream link, a local video file, a screen recording, a meeting recording, or a playlist/folder of videos — "watch this", "summarize this video", "what's in this recording". Use this to actuallysetup-watch-skill— One-shot setup for Watch Skill — installs the engine, bootstraps ffmpeg/yt-dlp, registers the MCP server in every detected AI agent (with backups), and configures the user's preferred vision provider. Run once after installing the plugin.
MCP servers (1)
watch-skill— stdio server
About this listing
Watch Skill is a community plugin on GitHub by oxbshw, written as a Claude Code plugin that Grok Build installs. We install-checked it in Grok; we haven't run it in a live session or reviewed its code. When we installed it, Grok found 11 skills, 1 MCP server. Its license is MIT. We checked commit f1317c8fe647. It's one of 44 listings in the Observability & Monitoring category on GrokStacks. It also appears in best Grok Build plugins for observability and monitoring.
Stacks Weekly: every new Grok Build and Codex plugin, every Tuesday.
One short email for GrokStacks and CodexStacks readers, with each new plugin's install line and what it ships. Subscribe and get the free starter list: 20 Grok Build plugins worth installing first.
You'll get a confirmation email first. Unsubscribe anytime.
Watch Skill FAQ
How do I install Watch Skill in Grok Build?
Run grok plugin install oxbshw/watch-skill --trust. We install-checked this command in Grok Build. We install this repo directly: Grok Build 1.0.62 can't install it through its own marketplace file. Then start a new Grok session, or press r in /plugins, to load it.
Is Watch Skill an official Grok Build plugin?
No. Watch Skill is a community plugin on GitHub by oxbshw, not in xAI's official Grok Build marketplace. We checked that it installs and what Grok loads from it, but haven't run it in a live Grok session or reviewed its code.
What does Watch Skill include?
According to what Grok loaded in our install check, Watch Skill includes 11 skills, 1 MCP server: asking-with-evidence, configuring-vision, extracting-structure, learning-from-mistakes, recovering-from-errors, sharing-results….
Details
- Type
- Plugin
- Category
- Observability & Monitoring
- Source
- Community Claude Code plugin, installs in Grok
- License
- MIT
- Author
- oxbshw
- Version
- 1.4.3
- Keywords
- agent-observability, agent-skills, agentic-ai, ai-agents, audio-understanding, computer-vision, deepseek, deepseek-harness, deepwatch, llm, local-first, mcp
- Source repo
- https://github.com/oxbshw/watch-skill
- Checked commit
f1317c8fe647- Homepage
- https://pypi.org/project/watch-skill/
- Checked by GrokStacks
- Install-checked in Grok Build on Oct 18, 2026; not run in a live session
Something wrong or missing? Submit or improve this listing. Marketplace listings also refresh automatically every night.
For maintainers: badge and embeds
Maintain Watch Skill? Add the badge; it links to this page.
Markdown
[](https://grokstacks.com/p/watch-skill/)HTML
<a href="https://grokstacks.com/p/watch-skill/"><img src="https://grokstacks.com/badge.svg" alt="Listed on GrokStacks" width="164" height="28"></a>Embed card (static HTML and inline CSS, no JavaScript). Preview:
<a href="https://grokstacks.com/p/watch-skill/" style="display:block;max-width:380px;box-sizing:border-box;padding:14px 16px;border:1px solid #2a2e3c;border-radius:12px;background:#0a0b0f;color:#e8eaf0;font:14px/1.45 -apple-system,Segoe UI,Roboto,Helvetica,Arial,sans-serif;text-decoration:none"><span style="display:block;font-size:16px;font-weight:700;color:#e8eaf0">Watch Skill</span><span style="display:block;margin:2px 0 8px;font-size:12px;color:#9aa0b1">Community plugin for Grok Build · 11 skills, 1 MCP server</span><span style="display:block;margin:0 0 10px;color:#c9cdd8">Give AI agents eyes, ears, and verifiable results. Watch Skill turns video, audio and screen activity into searchable…</span><span style="font-size:12px;font-weight:600;color:#7c9cff">View on GrokStacks →</span></a>Markdown summary
**[Watch Skill on GrokStacks](https://grokstacks.com/p/watch-skill/)**: Community plugin for Grok Build with 11 skills, 1 MCP server. Install: `grok plugin install oxbshw/watch-skill --trust`More on badges, embeds and keeping your listing up to date: GrokStacks for plugin authors.
Also in Observability & Monitoring
Agents Observe
Real-time observability of claude code sessions & multi-agents.
Claude Ops
Business operating system for Claude Code — 57 skills, 21 agents, smart daemon. Unified inbox (WhatsApp/Email/Slack/Telegram), autonomous PR merge, full-AWS monitoring, revenue (Stripe+RevenueCat), e-commerce (Shopify), marketing (Klaviyo/Meta/GA4), voice (Bland/ElevenLabs), APM (Datadog/NewRelic/OTEL), YOLO mode.
Ctrlb Decompose
LLM-ready reasoning surface over logs
DotNext
Next generation API for .NET
Eigenwise Toolshed
Claude Code plugins for the work that keeps coming back: repo maps, conditional rules, ticketed parallel work, extra subscription models, local usage metrics, and guided setup.
Emulo
Mine your Claude Code and Codex logs into a local you.md agent profile.