Clip the exact moments your viewers rewatch most
Your agent reads the most-replayed heatmap on your YouTube video and clips exactly the moments viewers keep rewatching — the audience picks the clips, and the AI score just breaks ties.
Run it with your agent
- Open Settings → Connectors in your client.
- Find OpusClip, click Add.
- Sign in with your OpusClip account in the OAuth window.
- Run
claude mcp add --transport http opusclip <url> - Start Claude Code and run
/mcp - Approve the OAuth sign-in with your OpusClip account
- Open Settings → MCP → Add new server in Cursor
- Paste the OpusClip MCP URL (Streamable HTTP)
- Sign in with OAuth on first use
- Open Settings → Connectors → Create in ChatGPT
- Paste the OpusClip MCP server URL
- Authenticate with OAuth using your OpusClip account
- Add OpusClip to
.vscode/mcp.json(type: http) - Open the MCP view in VS Code
- Sign in with OAuth when prompted
https://api.opus.pro/api/mcpWhat this workflow does
YouTube already tells you which moments of your video people can't stop rewatching — the most-replayed heatmap on every watch page. Most creators glance at it and move on. This workflow treats it as an instruction: the agent reads the heatmap's spikes, maps them to timestamps, and clips exactly those windows. Instead of predicting what might hook viewers, you're clipping what already did.
What you need
- An OpusClip account (MCP tool calls require a Pro plan)
- Claude, Cursor, or any MCP-enabled agent with the OpusClip MCP connected (api.opus.pro/api/mcp)
- A YouTube video with enough views for the most-replayed heatmap to appear (typically tens of thousands)
How it works
- Give the agent a video URL. Any of your videos where the heatmap is showing.
- It reads the replay spikes. The heatmap's peaks become a ranked list of timestamps, ordered by rewatch intensity.
- Each spike becomes a clip. The agent submits the video and cuts clips centered on those exact windows, with captions and vertical reframing.
- The AI score breaks ties, not the news. When spikes overlap or a window can be cut multiple ways, the virality score picks the stronger edit — but the audience signal decides what gets clipped at all.
- You get clips ranked by proven attention. Each one labeled with its spike rank, ready to post or schedule.
Try this prompt
Look at the most-replayed heatmap on this video of mine: [YouTube URL]. Find the top 5 replay spikes, clip a 30–60 second window around each one with captions and vertical reframe, and rank the results by rewatch intensity. Use the virality score only to choose between overlapping cuts.
Tips
- Merge spikes that sit within ~20 seconds of each other — they're usually one moment, not two.
- Compare the heatmap's picks against the AI's own top picks; where they agree, you've found your strongest clip.
- Run it on your back catalog's most-viewed videos first — older hits have the richest heatmaps.
