CLI commands

The Aucert CLI provides three commands in Phase 1: init, run, and status.

aucert init

Initialize a new Aucert project in the current directory. Creates an aucert.config.yaml with default test configuration.

Syntax

aucert init [--app <path>] [--platform <android|ios>] [--force]

Flags

Flag	Description	Default	Required
--app	Path to APK/AAB file	Auto-detected from ./app/build/outputs/	No
--platform	Target platform	android	No
--force	Overwrite existing config	false	No

Examples

Basic initialization — auto-detects your APK:

cd my-android-app
aucert init

Output

✓ Detected APK: ./app/build/outputs/apk/debug/app-debug.apk
✓ Platform: android
✓ Created aucert.config.yaml

Explicit APK path — useful when multiple build variants exist:

aucert init --app ./app/build/outputs/apk/release/app-release.apk

Re-initialize — overwrite an existing config:

aucert init --force

Exit codes

Code	Meaning
0	Config created successfully
1	Config already exists (use --force to overwrite)
2	No APK found and --app not specified

---

aucert run

Execute a test run against your application. This is the primary command — it triggers the full AI testing pipeline.

Syntax

aucert run [--scope <scope>] [--watch] [--output <format>] [--max-tests <n>]

Flags

Flag	Description	Default	Required
--scope	Test scope: navigation, user-flows, regression, or all	all	No
--watch	Stream results in real-time as tests complete	false	No
--output	Output format: json, junit, html	From config (default: json)	No
--max-tests	Maximum number of test scenarios to generate	From config (default: 100)	No

Examples

Run all tests — default behavior:

aucert run

Output

▶ Starting test run...
  Platform: android
  APK: ./app/build/outputs/apk/debug/app-debug.apk
  Scope: all (navigation, user-flows, regression)

⏳ Generating test scenarios... (12 scenarios)
⏳ Executing on emulator... [████████████████████] 12/12
⏳ Analyzing results...

✓ Test run complete
  Passed: 10/12
  Failed: 2/12
  Confidence: 94.2% average

Results saved to ./aucert-results/run-2026-04-07T10-30-00Z.json

Run with live streaming — see results as each test completes:

aucert run --watch

Output (streaming)

▶ [1/12] Login → Home navigation ........................ PASS (98.1%)
▶ [2/12] Login with invalid credentials ................. PASS (97.5%)
▶ [3/12] Search → Product detail → Cart ................ PASS (95.3%)
▶ [4/12] Cart → Checkout → Payment ..................... FAIL (72.1%)
  ⚠ Expected: payment form displayed
  ⚠ Actual: loading spinner persisted after 5s timeout
...

Navigation tests only — fastest scope, tests screen transitions:

aucert run --scope navigation

Generate JUnit XML — for CI/CD test summary integration:

aucert run --output junit

Output location

Results saved to ./aucert-results/run-2026-04-07T10-30-00Z.xml

Exit codes

Code	Meaning
0	All tests passed
1	One or more tests failed
2	Configuration error (missing config, invalid APK path)
3	Authentication error (missing or invalid API key)
4	Network error (cannot reach Aucert API)

tip

In CI/CD, exit code 1 (test failures) should fail the build. Exit codes 2–4 indicate setup problems that need fixing before tests can run.

---

aucert status

Check the status of current or past test runs. Useful for polling run completion in scripts or reviewing historical results.

Syntax

aucert status [--latest] [--run-id <id>] [--format <format>]

Flags

Flag	Description	Default	Required
--latest	Show the most recent run	false	No
--run-id	Show a specific run by ID	—	No
--format	Output format: table, json	table	No

Examples

Show all recent runs:

aucert status

Output

Run ID                               Status    Tests   Passed  Failed  Date
───────────────────────────────────────────────────────────────────────────────
a1b2c3d4-e5f6-7890-abcd-ef1234567890 complete  12      10      2       2026-04-07 10:30
f9e8d7c6-b5a4-3210-fedc-ba0987654321 complete  8       8       0       2026-04-06 14:15
11223344-5566-7788-99aa-bbccddeeff00 running   15      —       —       2026-04-07 11:00

Show the latest run:

aucert status --latest

Show a specific run with JSON output — useful for scripting:

aucert status --run-id a1b2c3d4-e5f6-7890-abcd-ef1234567890 --format json

Output

{
  "run_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "status": "complete",
  "started_at": "2026-04-07T10:30:00Z",
  "completed_at": "2026-04-07T10:32:15Z",
  "scenarios": 12,
  "passed": 10,
  "failed": 2,
  "average_confidence": 0.942
}

Exit codes

Code	Meaning
0	Status retrieved successfully
1	Run ID not found
3	Authentication error

---

Global flags

These flags work with any command:

Flag	Description
--help, -h	Show help for a command
--version, -v	Show CLI version
--verbose	Show debug-level output
--config <path>	Use a custom config file path (default: ./aucert.config.yaml)
--no-color	Disable colored output (auto-disabled in non-TTY environments)

Authentication

The CLI authenticates via API key. Two methods:

1. Environment variable (recommended for CI/CD):

   export AUCERT_API_KEY=your-api-key
   aucert run

2. Interactive login (recommended for local development):

   aucert auth login

   Stores the token in ~/.aucert/credentials.

What's next

• Configuration — Customize your aucert.config.yaml
• CI/CD integration — Automate with GitHub Actions
• Run your first test — Step-by-step tutorial