Docs

Scry documentation

How to connect over MCP, write read-only SQL against the archive, and use vectors — plus the schema, freshness, and provenance reference.

Start here

  • ConnectionScry is an MCP server at https://mcp.scry.io. Add the URL to Claude Code, Codex, ChatGPT, Claude.ai, or Cursor and approve the sign-in; per-client steps.
  • First queryThe agent calls scry_context once, scry_schema before SQL, then scry_query with a LIMIT. Rows come back with their sources.
  • HTTP APIWithout an MCP client, call GET /v1/scry/context, GET /v1/scry/schema, then POST /v1/scry/query with Content-Type: text/plain and a bearer key.

Reference

Connect Scry to ChatGPT, Claude, Codex, and other agents Add Scry as an MCP server in ChatGPT, Claude.ai, Claude Desktop and Cowork, Claude Code, Codex, Cursor, or any MCP client, then ask it structured questions over the corpus.HTTP API The Scry API is the MCP endpoint: plain HTTPS JSON-RPC, one URL, no client library. Raw HTTP routes ride underneath for services that want them.Turing-complete search Scry search computes: recursive SQL and fixpoint programs run unbounded iteration with conditional branching over the corpus, under budgets and deadlines you set. Graph walks, negation, whole-population aggregates — the answer, not a page of links.The calculus of search Everything in Scry is a relation, and a query constructs new relations from old ones. Conjunction, union, negation, aggregation, semantic ranking, and the fixpoint operator μ compose without ceiling — the small calculus behind programmatic search, with examples verified on the live engine.Search operators Use exact phrases, exclusions, source filters, author filters, dates, engagement floors, and language filters in Scry search.Rerank Order documents you already hold with the authenticated Scry rerank endpoint.Pricing: one meter, one multiplier The whole Scry price law in plain words: the machine's own rate, free while the box is at most half full and one smooth curve past it, posted rent when the box is contested, five ways to lock a price, bytes out, zero retention, budgets, and the limits that keep a query bounded.Errors Handle Scry HTTP error envelopes, status codes, retry guidance, and long-query failures.Vector search and compositional embeddings Use Scry vector helpers and the embeddings.* relations together with source-native SQL filters.Schema, provenance, and freshness Inspect Scry schema, source-native tables, provenance fields, freshness state, and query-visible corpus surfaces before running broad research queries.Agent setup for Scry Connect Claude Code, Codex, or another agent to Scry over MCP, or give it an API key for the HTTP API, and let it compose bounded SQL over public corpus sources.