L1HALO

Mission 01 ▸ Local research MCP Operational

Your GPU answers first

Claude Code hands codebase questions to a local 27B model, checks the citations, and takes back only what it gets wrong.

Eval score
70%
Claude Sonnet 5.5 ▸ 71%
Hallucination
13%
down from 26%
Claude tokens
0
per local answer
Mission file

MCP-01 ▸ Ground station ▸ one question, start to finish

  1. Claude sends a code question to the local model instead of reading the files itself.

  2. The model searches an index of 1,395 Lua files and reads only what it needs.

  3. The answer comes back with citations, each marked as read or unverified.

  4. Medium-confidence answers and Java engine questions go up the relay to Claude.