Preuve AI
Market research and competitive intelligence for founders, and the white-label engine agencies, accelerators and consultants use to pre-qualify inbound startups. Real signals, not a confident AI score with nothing behind it. A free scan in about 60 seconds, a full paid report in about ten minutes. Preuve AI uses 10 AI agents to scan 60+ live sources in parallel and return a verdict that has to be earned (median 54/100, only 17.5% reach 70), mapped competitors, demand signals, blind spots, and next steps. Every paid report runs on Claude Opus 5.5, Anthropic's most capable model.
Website: preuve.ai
Capabilities
- Preuve AI Developers (AI agent): Validate startup ideas from Claude, Cursor, or Codex. Viability score, competitors, source-linked evidence.
- Preuve Agent API (API): Market research and competitive intelligence for startup ideas, built for agents and scripts: a verdict that has to be earned (median 54/100, 17.5% reach 70),…
- Preuve AI Developers (MCP server): This Model Context Protocol server provides search and retrieval tools for the Preuve AI Developers site. Use it to answer questions from public site content.…
- Preuve MCP server (MCP server):
- Preuve AI (Website): Market research and competitive intelligence for founders, and the white-label engine agencies, accelerators and consultants use to pre-qualify inbound…
- Preuve AI Developers (Website): Validate startup ideas from Claude, Cursor, or Codex. Viability score, competitors, source-linked evidence.