Summary
A remote MCP server that runs full multimodal analysis on video URLs from YouTube, TikTok, Instagram, Vimeo, Twitter, and direct links. It exposes four tools: quick_transcribe for timestamped audio with speaker ID, deep_analyze for the full pipeline (transcript plus keyframe vision plus OCR in one structured output), clip_context for analyzing specific timestamp ranges, and batch_analyze for processing up to 10 videos in parallel. Connects over streamable HTTP with OAuth, no local installation. Built on yt-dlp, Groq Whisper, Tesseract OCR, and Claude Vision. Reach for it when you need an agent to reason over what's shown on screen, not just what's said in the audio.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Tools
Public tool metadata for what this MCP can expose to an agent.
4 tools
quick_transcribeTRANSCRIPTION ONLY (no visual analysis). For vision, OCR, charts, or keyframe analysis, use deep_analyze instead. Returns timestamped transcript with speaker labels. Supports YouTube, Instagram Reels, Vimeo, Twitter/X, TikTok, and direct video URLs. Costs 1 credit.2 params
TRANSCRIPTION ONLY (no visual analysis). For vision, OCR, charts, or keyframe analysis, use deep_analyze instead. Returns timestamped transcript with speaker labels. Supports YouTube, Instagram Reels, Vimeo, Twitter/X, TikTok, and direct video URLs. Costs 1 credit.
Parameters* required
languagestringISO 639-1 language code (e.g., 'en', 'es', 'ja'). If omitted, language is auto-detected from the audio.default: auto
video_urlstringPublic URL of the video to transcribe. Supports YouTube, Instagram Reels, Vimeo, Twitter/X, TikTok, and direct video URLs (MP4, WebM, MOV).
deep_analyzePRIMARY tool for video understanding. Full multimodal pipeline: transcript + visual keyframe analysis + OCR + chart/diagram extraction. Returns unified analysis with Summary, Key Claims, Visual Assets, Data Extracted, Entities Mentioned. Costs 5 credits.2 params
PRIMARY tool for video understanding. Full multimodal pipeline: transcript + visual keyframe analysis + OCR + chart/diagram extraction. Returns unified analysis with Summary, Key Claims, Visual Assets, Data Extracted, Entities Mentioned. Costs 5 credits.
Parameters* required
focusstringAnalysis focus area: general (default), technical, or crypto-specific.one of
general · technical · cryptodefault: generalvideo_urlstringPublic URL of the video to analyze with the full multimodal pipeline (transcription + keyframe vision + OCR).
clip_contextAnalyze a specific segment of a video by timestamp range. Default is full multimodal analysis (transcript + vision + OCR, 3 credits). Pass mode='quick' for transcript-only (1 credit). Use when you only need a section, not the full video.4 params
Analyze a specific segment of a video by timestamp range. Default is full multimodal analysis (transcript + vision + OCR, 3 credits). Pass mode='quick' for transcript-only (1 credit). Use when you only need a section, not the full video.
Parameters* required
modestringAnalysis mode. 'deep' (DEFAULT) returns full multimodal analysis — transcript + vision + OCR (3 credits). 'quick' returns transcription only (1 credit).one of
quick · deepdefault: deepend_timestringClip end timestamp. Same format as start_time. Clip length must be between 5 seconds and 30 minutes.
video_urlstringPublic URL of the video. Only the specified time range is downloaded and analyzed.
start_timestringClip start timestamp. Accepts 'MM:SS', 'HH:MM:SS', or seconds as a string (e.g., '870').
batch_analyzeProcess multiple videos and get cross-video synthesis. Max 10 URLs. Returns individual results plus common themes, entity overlap, and contradictions. Credit cost is per-video rate with 10% discount on 5+ videos.2 params
Process multiple videos and get cross-video synthesis. Max 10 URLs. Returns individual results plus common themes, entity overlap, and contradictions. Credit cost is per-video rate with 10% discount on 5+ videos.
Parameters* required
modestringPer-video analysis mode. 'quick' (default) returns transcription only (1 credit each) — appropriate for bulk transcription. 'deep' returns full multimodal (5 credits each) — for vision/OCR across multiple videos. 10% discount applied at 5+ URLs.one of
quick · deepdefault: quickvideo_urlsarrayArray of video URLs to analyze in parallel. Max 10 URLs per call. Returns individual results plus cross-video synthesis (common themes, shared entities, contradictions).
Contendeo
Give your agent eyes.
Contendeo is the multimodal layer that lets your AI actually see video — not just read its transcript.
What it is
Contendeo is a remote MCP server that gives LLMs frame-level context from video — transcription, keyframe vision analysis, OCR, and structured output, unified into a single response your agent can reason over.
Paste a YouTube, Instagram Reels, Vimeo, Twitter/X, TikTok, or direct video URL into Claude (or any MCP client). Contendeo downloads, transcribes, extracts keyframes, runs OCR, analyzes visuals, and returns a structured document.
Live at contendeo.app. MCP endpoint at contendeo.app/mcp/.
Why it exists
Transcripts capture what was said. They don't capture what was shown — chart values, UI states, overlays, code on screen, circled regions, dashboard numbers. For any video where the visual layer carries information (trading tutorials, product demos, technical walkthroughs, data dashboards, design reviews), transcript-only analysis misses half the signal.
Contendeo closes that gap.
See the side-by-side comparison at contendeo.app/demo.
Tools
Four tools exposed over MCP. Credit costs charged per successful call. Cache hits are free.
| Tool | Credits | What it does |
|---|---|---|
quick_transcribe | 1 | Timestamped transcript with speaker identification. Audio-only. Fast. |
deep_analyze | 5 | Full multimodal pipeline — transcript + keyframe vision + OCR, unified output. |
clip_context | 1 (quick) / 3 (deep) | Analyze a specific timestamp range without paying for the full video. |
batch_analyze | per-video, −10% at 5+ | Process up to 10 videos in parallel with cross-video synthesis. |
Full schemas and response formats: docs/tools.md.
Installation
Contendeo is a remote MCP server — no local install, no package download. Connect via URL from your MCP client.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"contendeo": {
"url": "https://contendeo.app/mcp/"
}
}
}
Claude will walk you through OAuth on first use.
Claude Code
claude mcp add --transport http contendeo https://contendeo.app/mcp/
Other MCP clients (Cursor, Continue, Cline, etc.)
Any client that supports remote MCP servers over streamable HTTP. Point it at https://contendeo.app/mcp/.
Full walkthrough with screenshots: docs/installation.md.
Authentication
Contendeo uses OAuth 2.0. First use flow:
- Your MCP client initiates a connection
- Browser redirect to contendeo.app sign-in (Google OAuth or email)
- Approve the MCP client
- Bearer token issued, client reconnects, tools become available
New accounts get 10 free credits. No card required.
Details: docs/authentication.md.
Pricing
| Plan | Price | Credits | Notes |
|---|---|---|---|
| Free | $0 | 10 on signup | All 4 tools, no card |
| Pro | $12/mo (₹999) | 100/mo | Priority queue, rollover to 200 |
| Power | $39/mo (₹3,299) | 500/mo | Batch, webhooks, rollover to 1000 |
| PAYG | $0.15/credit | on demand | No subscription |
Cache hits are free. Failed analyses auto-refund.
Details: docs/pricing.md.
Stack
Under the hood:
- yt-dlp — video download across 1000+ sites
- Apify — Instagram Reels fetching
- Webshare — residential proxy pool for YouTube anti-bot
- ffmpeg — audio extraction + keyframe scene detection
- Groq Whisper (large-v3) — transcription
- Tesseract OCR — deterministic text extraction from frames
- Claude Vision (Sonnet 4.6) — visual frame analysis
- Supabase — auth + Postgres
- Redis — URL-hash result cache
- FastAPI + MCP Python SDK — server, streamable HTTP transport
- Traefik — SSL termination + routing
Links
- Landing — contendeo.app
- Demo — contendeo.app/demo
- MCP endpoint — contendeo.app/mcp/
- Dashboard — contendeo.app/dashboard
- Docs —
docs/ - Server card — contendeo.app/.well-known/mcp/server-card.json
Support
Questions, bug reports, feature requests: open an issue.
Commercial/partnership inquiries: @0xKaroshi on X.
License
MIT — covers this wrapper repo (documentation, examples, manifest). The production server source is not open-sourced.
See LICENSE.
Contendeo — every frame, every word, every edge.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →