Leads

AI-powered lead qualifying from any data source.

• The HTTP endpoint at main.ts ingests all inbound POST request data and passes it to the OpenAI agent in agent.ts.
• The agent uses the lead data from your POST request and its web search tool to determine if the lead is a match for the criteria in PROMPT.txt.
• By default, the AI returns:

  {
    "name": "(normalized name)",
    "match": "(true or false)",
    "reasoning": "(explanation)"
  }

• Every lead is added to the leads SQLite table with the columns:
  • id — auto-incremented, starts at 1
  • timestamp — Unix timestamp
  • input_data — JSON of your POST body
  • output_data — JSON of AI response
• The main.ts dashboard shows a history of all lead assessments, successful matches first. Clicking any lead shows the full inbound and output data from the assessment.

1. Click Remix
2. Save your OPENAI_API_KEY as an environment variable
3. Customize PROMPT.txt (Don't remove any fields from the structured response)
4. Get the val's HTTP endpoint from main.ts:
5. Start forwarding leads to this webhook from any source as a POST request.
6. That's it! You now have a growing leads SQLite table with one column for your original input_data and a second column for the AI evaluator's output_data. You also have a dashboard of this data hosted at the main.ts val endpoint.

• You can add any number of fields to the AI's output instructions in PROMPT.txt and they'll be added to output_data. You can use this to expand the AI's assessment, enrich your leads, or even just normalize lead data coming from different sources.
• You can also add query parameters when you POST to the val endpoint. (ex: your-val-endpoint.val.run?source=signup-page). The parameters will be bundled in the _query object of input_data. This can be useful for tracking where a lead was forwarded from.