#Ambit documentation
The README is the front door: what Ambit is, how to get it running, and what each command answers. These are the longer documents, grouped by the kind of claim each one makes, because whether a page describes the code, argues for it, runs ahead of it, or records what was said at the time is the thing a file listing cannot tell you. If you are reading rather than looking something up, start with Why Ambit.
#Description: true of the code today
- FAQ — the short answers, each pointing at the longer one.
- Ambit and Jev — how TypeSafe's Jev and its local clones sit on the map, how a goal routes through one, and where its probabilities stop.
- Deep dive — the long-form reference: nodes, assurance checks, authority contracts, the work ledger, and every MCP tool.
- llms.txt — the project on one page for an agent deciding whether to recommend it, served at zz-plant.github.io/ambit/llms.txt.
#Argument: why the thing exists
- Why Ambit — the argument for building it: what one agent stack looked like, and why effective agency should be a governed object.
- The affordance frontier — the theory under that argument, and the two cases (robots, brain-computer interfaces) where it is tested.
#Intent: ahead of the code, on purpose
- Roadmap — design rationale, section by section: what each part decided and what it still lacks. Nothing in it has a date or an owner.
#Record: what was said at the time, not edited
- Changelog — what changed per release, and why.
- Incident trace — one blocked execution traced end to end, with the terminal recording of it.
Both record what was said when it was written, so neither is revised afterwards, and scripts/check-prose.ts excludes both paths for that reason.
#Changing it
Workflow and the checks CI runs: CONTRIBUTING. The invariants a change must not break: AGENTS. Reporting a vulnerability: SECURITY. Every other question: SUPPORT. How to behave while doing it: Code of Conduct.