| Reference |
What this section is, which pages it has, and which dataset it is showing. |
| Configuration |
Every setting Gantry reads, its default and the tiers that can set it, and every GANTRY_* environment variable. |
| Command line |
Every verb the binary answers to and every flag it parses, with the builds that have them. |
| Orchestration patterns |
Every pattern Gantry ships — what each declares, what it takes, and its whole text as the annotated program it already is. |
| Agent prompts |
Every prompt Gantry hands its agents, which stage of a run gets which, and where a copy of one has to go to displace it. |
| Harnesses and roles |
Every agent harness Gantry drives, the environment keys that configure each one, and which role runs which stage of the pipeline. |
| MCP tools |
Every tool the run-scoped MCP server advertises, its arguments, and whether it reads the run or acts on it. |
| Menus and keys |
Every menu the monitor has with its rows in order, and every key it binds, grouped as the in-app help overlay groups them. |
| When a run stops |
The closed sets an operator meets when a run stops or asks a question: outcomes, dispositions, ledger statuses, prompt kinds, run states and exit codes. |
| Journal events |
Every event the run journal and the headless stream write, with the name each record carries on the wire. |