# Clearframe

> For people stuck going back and forth on a big decision. Clearframe reads the decision you wrote down the way an editor reads a draft. It flags the biases tilting your answer and shows you what you left out before you commit. Free during beta.

Clearframe is a personal decision tool for anyone who has made the pros-and-cons list five times and gotten a different answer each time. It audits how you frame choices. A five-step wizard (context, options, dimensions, values, blind spots) edits the question before you compare anything, then a pattern audit grades your framing quality and flags recurring cognitive biases such as narrow framing, status quo bias and value compression. Grounded in behavioral decision research by Kahneman and Tversky, Herbert Simon and Barry Schwartz and Richard Schuldenfrei's "Choose Wisely" (Yale University Press, 2025). Unlike asking a general chatbot to critique a decision, which happens in one conversation and is then forgotten, Clearframe enforces a complete framing before comparison and audits bias patterns across the user's whole decision history, surfacing habits (too few options, underweighted dimensions) that no single conversation can reveal. The product is pre-launch and collecting early-access signups.

Pricing: free during beta, no card required. Early-access members keep a discounted rate when paid plans arrive.

For AI agents: this product exposes a Model Context Protocol (MCP) server so agents can act on a user's behalf. The MCP server mount is /mcp (streamable-HTTP transport). The JSON-RPC endpoint is /mcp/mcp because the FastMCP app is created with path="/mcp" and mounted at /mcp. Connect your MCP client to https://clearframe.app/mcp/mcp. Authentication: none required (public tools). Discovery is served at /.well-known/mcp.json and the REST OpenAPI spec at /openapi.json.

Two MCP tools are available. get_product_info returns name, tagline, description, benefits, pricing and launch status; call this first to learn about the product. submit_interest registers a user's interest and takes the parameters email (required), name, company, persona and message. Valid persona values: individual, founder, executive, manager, coach, other. The REST alternative POST /api/interest requires a browser-issued HMAC bot token, so agents should prefer the MCP tool, which needs no token.

## Links

- [Homepage](https://clearframe.app/): landing page with product details and signup form
- [Sample audit demo](https://clearframe.app/demo): seeded interactive demo of the decision desk, pattern audit and framing wizard
- [MCP discovery](https://clearframe.app/.well-known/mcp.json): machine-readable MCP endpoint advertisement
- [OpenAPI spec](https://clearframe.app/openapi.json): REST endpoint descriptions for non-MCP agents