`poisson-prove runs`

Lists every run in the store, newest first, with its number, name, id, status, backend, target and row counts, so the next command can name a run. Any of the three works wherever a command takes a run: the name ("mock-agent 2026-10-26 a"), the id, or the number (#1 is the newest complete run). So do latest and previous, the two newest complete runs, and latest:TARGET, the newest complete run of one target. A partial run has no number and is never latest; name it by its name or id.

Help

poisson-prove runs --help

Output17 lines

usage: poisson-prove runs [-h] [--json] [--store PATH] [--plain] [--debug]

List every run in the store with its status, backend, target and row counts,
so the next command can name one.

options:
  -h, --help    show this help message and exit
  --json        print the list as one JSON document instead of text
  --store PATH  the store folder (default: $POISSON_PROVE_STORE when set, else
                ./prove-runs)
  --plain       print plain text, as a pipe gets it, on a terminal too
                (default: pretty on a terminal, plain otherwise)
  --debug       also print the traceback of an unexpected failure

examples:
  poisson-prove runs
  poisson-prove runs --json

Options

• --json prints the list as one JSON document, poisson-prove.runs/1, each run with its run_id, run_name, status and counts.
• --store PATH lists another store than ./prove-runs (or $POISSON_PROVE_STORE); --plain prints plain text on a terminal too; --debug adds the traceback of an unexpected failure.

Examples

The example’s two runs:

poisson-prove runs

Output4 lines

ok  2 runs in prove-runs
#  name                     id                     status    backend  target      rows   rows without an outcome  created
1  mock-agent 2026-10-26 b  20261026T144648Z-0eef  complete  local    mock-agent  1,746  0                        2026-10-26T14:46:48Z
2  mock-agent 2026-10-26 a  20261026T144646Z-ba84  complete  local    mock-agent  1,746  0                        2026-10-26T14:46:46Z

rows without an outcome counts the episodes that errored, those stopped at --episode-timeout included; inspect shows them by their error. A complete run in which every episode errored is shown as stopped, as the sweep showed it; runs --json and the run’s manifest keep complete, with rows and no_outcome carrying the fact. The same list as JSON:

poisson-prove runs --json

Output7 lines

{
  "schema": "poisson-prove.runs/1",
  "verdict": "ok",
  "store": "prove-runs",
  "runs": [{"run_id": "20261026T144648Z-0eef", "run_name": "mock-agent 2026-10-26 b", "status": "complete"}]
  …
}

A store that does not exist is refused, with the command that would make one:

poisson-prove --store nowhere runs  # exits 1

Output2 lines

error: no store at nowhere. …
  next: poisson-prove --store nowhere sweep SPEC.toml --target pkg.module:rollout

Exit codes

• 0: the list was printed; an empty store is listed as empty.
• 1: refused: no store at that path, or a store this version cannot read.
• 2: the command line is wrong.
• 3: an unexpected failure; --debug prints its traceback.
• 130: interrupted with Ctrl-C.

What it writes

Nothing.