Creating Test Runs

Creating Test Runs in Testmo

scripts/testmo_create_runs.py selects the cases each filter spec matches and, only with --create, creates one run per spec through the Testmo API.

Writes mutate the live Testmo instance. Always dry-run and have the user review before --create.

Prerequisites

• TESTMO_API_KEY exported in the environment. Reference it only by variable — never print, echo, log, commit, or pass the value as a command-line argument.
• python3 on PATH. curl is used automatically as a fallback on TLS-inspecting networks; see references/tls-inspecting-proxies.md (opens in a new tab) only if that fallback needs explaining.

Paths

The script ships inside the installed plugin, so always address it through ${CLAUDE_PLUGIN_ROOT}. Shell state does not carry over between commands, so set the shorthand in every command:

SKILL="${CLAUDE_PLUGIN_ROOT}/skills/creating-test-runs"

Specs can live anywhere; pass their real paths. Never write a user’s spec into $SKILL/ — that is the plugin cache, and /plugin update overwrites it.

Workflow

1. Get the specs. Use the spec paths you were given, or write a new spec starting from examples/test-run.template.json (opens in a new tab), saved outside the plugin. Read references/api-reference.md (opens in a new tab) for the ids behind field values before writing or debugging a spec.
2. Confirm the milestone exists. The API cannot create milestones; the user creates them in the Testmo UI. Link runs with --milestone-name "<exact name>" or --milestone-id <id>.
3. Dry-run. Pass --spec once per run:

   bashCopy

   SKILL="${CLAUDE_PLUGIN_ROOT}/skills/creating-test-runs"
   python3 "$SKILL/scripts/testmo_create_runs.py" \
     --spec /path/to/a.json --spec /path/to/b.json \
     --milestone-name "<exact milestone name>"

   A single spec also prints its folder resolution, sample cases, and run payload.
4. Show the user the dry-run output and stop. Point out every row whose STATUS is not ready, the Existing line, and any case count that looks off (a large swing from last cycle usually means a renamed folder or tag). Wait for explicit confirmation. Never run the dry-run and --create in the same turn.
5. Create by re-running the identical command with --create.
6. Report the result, including every run listed under MANUAL STEP 2 REQUIRED — those runs are unfinished until the user does the UI pass in references/multi-configuration-runs.md (opens in a new tab).

Reading the dry-run

--create refuses the whole batch unless every row is ready. The Existing line reads NOT CHECKED without a milestone, and CANNOT CHECK when the API hides milestone links — in that case have the user check the milestone in the Testmo UI before creating.

If some creates fail, the script names them and exits non-zero. Re-run with only the failed specs; the others already exist and would now report DUPLICATE.

Filter-spec schema

Top-level keys describe the run; keys under filters select its cases.

{
  "project_id": 1,
  "run_name": "Password Manager",
  "run_state_id": 7,
  "tags": ["regression"],
  "filters": {
    "folder_paths": ["Web > Password Manager"],
    "test_type_ids": [19, 20],
    "case_state_ids": [4],
    "exclude_automation_type_ids": [10, 23, 24]
  }
}

Top level: project_id and run_name are required; every spec in one invocation must share the same project_id. Optional: run_state_id (default 7, In progress), milestone_id (overridden by the milestone flags), config_id (only for same-named runs split by Testmo Configuration), tags (labels the run — does not select cases), note.

Under filters every key is optional; keys are ANDed, values within a key are ORed:

• tags — select cases carrying these tag names or ids. Prefer ids when a name is duplicated.
• folder_paths / folder_ids — folders and their descendants (include_subfolders: false for exact).
• test_type_ids, team_ids — the case’s set must intersect the list.
• case_state_ids — e.g. [4] for Active.
• automation_type_ids / exclude_automation_type_ids — for manual-only, exclude [10, 23, 24].
• has_automation — bool.

A run defined purely by a tag must set filters.tags; setting only the top-level tags selects nothing. An empty or missing filters block matches every case in the project, so always check the dry-run count.

Keep run_name to the bare area (e.g. "Password Manager"); the release comes from the milestone and the platform from config_id.