About Hexum
Last updated 29 August 2026
Hexum is architecture control for coding agents. Deterministic hexum_check over MCP. Keep the editor you already use. Operator
of hexum.dev. Contact: how to reach Hexum.
FAQ
What is Hexum?
Hexum is an MCP server for Cursor, Claude Code, and other coding agents. It sits between the agent and the model: hexum_shrink sends this turn instead of the repository, hexum_check fails the turn when architecture.hexum is broken, and you keep your editor. On MBPP, context dropped about 85% with the same tests passing.
Does Hexum replace Cursor or Claude Code?
No. Hexum is a hosted MCP server. You keep Cursor, Claude Code, Antigravity, Codex, or Grok. Hexum is not an IDE and not a model.
How does Hexum reduce tokens?
The agent calls hexum_shrink before it reads widely. The model is sent this turn, not the repository. On MBPP, measured context dropped about 85% with the same tests passing. Example turn: 18.2k → 3.1k tokens, 41 files → 6. See /benchmarks and /methodology.
Is hexum_check an LLM?
No. hexum_check is deterministic. It maps files to layers in architecture.hexum and reports forbidden imports. It does not call a model.
How is Hexum billed?
Hexum bills tokens saved: input minus what still goes to the model. Cache hits are free. No monthly cap. No seat tax. Usage early access from €0.30 per million. Teams from €0.35 per million. See /pricing.
What is tokens saved?
Tokens in minus tokens out, when that is positive. Stripe is billed on tokens saved. Calls that do not save tokens still increment calls and the in/out counters.