Browser-local submitted JSON

Claim-to-evidence coverage workbench

Build a bounded coverage graph from exact IDs and caller-authored relations. Contradiction is a submitted label, missing evidence stays visible, and no model, embedding, reranker, fetch, or semantic judgment is used.

Explicit-submit workbench

Run submitted JSON locally

Parsing and transformation happen in this browser. Nothing runs when a sample loads, and no input is fetched, uploaded, stored, or retained.

Submitted JSONUse strict JSON matching the published tool schema.

Deterministic resultCopy and download unlock only for the current successful result.

Load the inert sample or enter JSON, then explicitly run.

Hosted API and MCP parity

The same tool ID is available through POST /api/run and the fixed MCP run_tool method. Hosted requests are response-only. This browser workbench does not call either endpoint.

{
  "tool_id": "claim-evidence-coverage",
  "input": "<SCHEMA_VALID_INPUT>",
  "response_mode": "full"
}