Usage
Subcommands
list supports the global --limit, --page, and --json flags — see Configuration → Global flags.
usage — token consumption
machina org usage answers “how many LLM tokens did this org consume”. Token usage is recorded on agent executions (execution_tokens), and the Client-API totals are page-level only, so the command iterates the org’s projects, paginates + sums client-side, and breaks the total down by project, agent, and day — surfacing the prompt/completion split (the cost-shape signal).
Unreachable / undeployed projects are listed under
projects_unreachable rather than failing the run; a project that is reachable but errors mid-scan flags the run INCOMPLETE / PARTIAL (a partial total is never presented as if it were whole).
