fabio-rovai/brain-in-the-fish
Evaluate anything. Predict everything. Hallucinate nothing. — SNN-verified document evaluation & prediction credibility MCP server.
<p align="center"> </p>
<h1 align="center">Brain in the Fish</h1>
<p align="center"> <strong>Evaluate anything. Predict everything. Hallucinate nothing.</strong> <br> <em>Evidence-verified document evaluation & prediction credibility — the brain that MiroFish was missing.</em> </p>
<p align="center"> </p>
<p align="center"> <a href="README.md">English</a> | <a href="README-CN.md">中文</a> | <a href="README-JP.md">日本語</a> </p>
---
<p align="center"> <br><em>Hierarchical knowledge graph — document structure, evaluation criteria, agent panel, and scoring connected in one tree</em> </p>
<p align="center"> <br><em>Detail panel showing ontology reasoning — what the node is, its structure, and why it exists in the knowledge graph</em> </p>
<p align="center"> <br><em>Evidence node inspection — properties, ontology role, connections, and provenance</em> </p>
---
A Rust MCP server that evaluates any document against any criteria using Claude subagents, with an Evidence Density Scorer (EDS) that makes hallucination mathematically detectable. Feed it a PDF and an intent — it returns structured scores, weakness analysis, and a full audit trail. Document evaluation is the core differentiator: BITF score
Loading reviews...