An open invitation to every AI lab.

ailetterbox is a public directory of human experts. MCP-native. Building the LTM — and we’d rather build it with you.

Frontier AI is fluent, fast, autonomous. What it lacks is a way to know when its output can be trusted.

Not “someone checked.” A human expert, verified, on the record. Anyone can look it up. Nobody can edit it.

That’s the trust artifact the industry currently lacks.

Verification only works where AI happens — inside the chats people already use.

Collaboration with the MCP-compatible labs — Anthropic, OpenAI, the frontier open-source community — is how ailetterbox becomes the default verify signal across AI.

This isn’t about competing with the model. It’s about giving the model somewhere trustworthy to point.

MCP-first. Any MCP-native runtime — Claude Desktop, Cursor, Zed, Windsurf, Continue.dev — calls mcp.ailetterbox.com. Four tools today: verify an expert’s address, search the directory, get a full profile, send a brief. AI-Chats routed by humans are no-fee today; autonomous AI-Agents launch with pricing in our FAQ.

A way to make ailetterbox available inside the AI experience your users already have. First-party integration, featured MCP surface, co-branded pilot — whatever fits.