How to Blend Movie Taste with OpenClaw
Use OpenClaw and ToolRouter to generate normalized multi-seed movie recommendation shortlists.
Tool
OpenClaw is the best fit when multi-seed recommendation runs need to stay normalized across many briefs. Use `blend_taste` for batch-style curation where the same schema should work for every taste profile.
Connect ToolRouter to OpenClaw
1Install the CLI
npm install -g toolrouter-mcp2Call tools directly from OpenClaw
toolrouter-mcp call web-search search --query "AI tools"
toolrouter-mcp toolsSteps
Once connected (see setup above), use the Similar Movies tool:
- Define the seed movie bundle, the avoid list, and the output shape before running the blend.
- Run `similar-movies.blend_taste` and keep the matched-seed and score fields consistent across briefs.
- Review the shortlist, then rerun only when you need deeper pagination or a stricter quality floor.
- Save the normalized output for repeatable recommendation workflows.
Example Prompt
Try this with OpenClaw using the Similar Movies tool
Use similar-movies to blend Zodiac, Gone Girl, and Knives Out into one shortlist, exclude Se7en, and keep the output normalized so I can compare recommendation bundles later.
Tips
- Stay within two to five seeds or the blend becomes noisy.
- Keep thresholds consistent if the output will be compared across runs.
- Use the matched-seed field as the first sanity check on every recommendation.