> ## Documentation Index > Fetch the complete documentation index at: https://docs.stably.ai/llms.txt > Use this file to discover all available pages before exploring further. # Run History (stably runs) > Browse and inspect test run history from the command line `stably runs` lets you browse test run history directly from the terminal — filter by branch, status, source, and more. Use it to investigate failures, check CI results, or pipe structured JSON into scripts and AI agents. ```bash theme={null} # List recent runs stably runs list # Filter by branch and status stably runs list --branch main --status failed # View details for a specific run stably runs view ``` ## List Runs `stably runs list` displays a table of recent test runs, sorted by most recent first. ```bash theme={null} stably runs list [options] ``` ### Options | Option | Description | | ---------------------- | ------------------------------------------------------------------------------------------------- | | `-b, --branch ` | Filter by branch name | | `-s, --status ` | Filter by status: `queued`, `running`, `passed`, `failed`, `timedout`, `cancelled`, `interrupted` | | `--source ` | Filter by source: `local`, `ci`, `web` | | `--trigger ` | Filter by trigger: `manual`, `scheduled`, `ui`, `api`, `github_action`, `suite_run` | | `--suite ` | Filter by suite name | | `-n, --limit ` | Number of runs to show (default: 20, max: 100) | | `--after ` | Show runs newer than this run ID | | `--before ` | Show runs older than this run ID | | `--json` | Output as JSON | ### Example Output ``` Run ID Status Branch Started Tests abc123def456 FAILED feature/checkout 2025-06-15 10:30 50 789xyz012345 PASSED feature/checkout 2025-06-15 09:15 50 fedcba987654 PASSED main 2025-06-14 22:00 48 Next page: stably runs list --before fedcba987654 ``` ### Filtering Combine filters to narrow results: ```bash theme={null} # Failed runs on a specific branch stably runs list --branch main --status failed # CI runs only stably runs list --source ci # Runs triggered by GitHub Actions stably runs list --trigger github_action # Failed runs from a specific suite, limited to 5 stably runs list --suite chromium --status failed --limit 5 ``` ### Pagination Results are sorted newest-first. When more results exist, a pagination hint is displayed: ```bash theme={null} # First page stably runs list --limit 10 # Next page (use the run ID from the hint) stably runs list --limit 10 --before ``` Use `--after ` to see runs newer than a given run, or `--before ` to see older runs. ## View Run Details `stably runs view` shows detailed information for a specific run, including test results and error messages. ```bash theme={null} stably runs view [options] ``` | Option | Description | | -------- | -------------- | | `--json` | Output as JSON | ### Example Output ``` Run abc123def456 Status: FAILED Branch: feature/checkout Commit: a1b2c3d Started: 2025-06-15 10:30 Duration: 2m 45s Source: ci URL: https://app.stably.ai/project/proj_xxx/playwright/history/abc123def456 Test Results: 47 passed, 2 failed, 1 flaky Failed/Flaky Tests: FAILED tests/checkout.spec.ts:42 - should complete purchase AssertionError: expected "Cart" to equal "Checkout" FAILED tests/payment.spec.ts:15 - should process payment (3 attempts) TimeoutError: page.waitForURL timeout 30000ms exceeded FLAKY tests/login.spec.ts:10 - should login successfully (2 attempts) TimeoutError: page.waitForSelector timeout 5000ms exceeded ``` The output includes: * **Run metadata** — status, branch, commit SHA, timing, source, and a clickable URL to the web dashboard * **Test results summary** — counts for passed, failed, flaky, timedout, skipped, and interrupted tests * **Failed/flaky test details** — file location, test name, attempt count (for retried tests), and error message from the last attempt Click the **URL** in the output to open the run in the Stably dashboard, where you can view traces, screenshots, and step-by-step execution for each test. ## JSON Output Use `--json` for structured output suitable for scripting, piping, or AI agent consumption: ```bash theme={null} # List runs as JSON stably runs list --branch main --status failed --json # View run details as JSON stably runs view --json # Pipe to jq for processing stably runs list --json | jq '.runs[] | select(.status == "FAILED") | .runId' ``` The JSON output passes the API response directly — date fields are ISO 8601 strings and null values are preserved as `null`. ```json theme={null} { "runs": [ { "runId": "abc123def456", "suiteName": "chromium", "status": "FAILED", "startedAt": "2025-06-15T14:30:00Z", "finishedAt": "2025-06-15T14:32:45Z", "durationMs": 165000, "branchName": "feature/checkout", "commitSha": "a1b2c3d4e5f6", "totalCount": 50, "trigger": "GITHUB_ACTION", "source": "CI", "autoHealStatus": "COMPLETED" } ], "hasMore": true } ``` ```json theme={null} { "runId": "abc123def456", "projectId": "proj_xxx", "suiteName": "chromium", "status": "FAILED", "startedAt": "2025-06-15T14:30:00Z", "finishedAt": "2025-06-15T14:32:45Z", "durationMs": 165000, "branchName": "feature/checkout", "commitSha": "a1b2c3d4e5f6", "trigger": "GITHUB_ACTION", "source": "CI", "passedCount": 47, "failedCount": 2, "flakyCount": 1, "timedoutCount": 0, "skippedCount": 0, "interruptedCount": 0, "totalCount": 50, "autoHealStatus": "COMPLETED", "autoHealReport": { "summary": "Fixed 1 of 2 failing tests", "issues": [ { "result": "fixed", "issueType": "test", "rootCause": "Selector changed after UI update", "fixApplied": "Updated selector to data-testid", "relatedTestCaseIds": ["tc-1"] } ] }, "testCases": [ { "testIdentifier": "tc-1", "title": "should complete purchase", "location": "tests/checkout.spec.ts:42", "status": "FAILED", "durationMs": 12000, "attemptCount": 1, "attempts": [ { "attemptNumber": 1, "status": "FAILED", "errorMessage": "AssertionError: expected \"Cart\" to equal \"Checkout\"", "errorType": "AssertionError" } ] } ] } ``` ## Use with `stably fix` `stably runs` pairs naturally with `stably fix` for investigating and repairing test failures: ```bash theme={null} # Find the failing run stably runs list --branch feature/checkout --status failed # Inspect it stably runs view # Fix it stably fix ``` The `stably fix` agent also uses run history internally to classify failures as regressions, pre-existing issues, or already-fixed problems — improving fix accuracy by understanding the broader context of each failure. ## Next Steps Automatically diagnose and fix failing tests Run tests locally or in CI with `stably test`