Exploratory Test

Execute multi-level exploratory testing that goes beyond smoke testing to actively find bugs through escalating test scenarios.

Task Tracking

At the start, use update_plan to track each step, restating any remaining steps of a parent workflow alongside them:

1. Load or create test plan
2. Determine testing approach
3. Run $user-experience skill (when user-facing)
4. Execute tests by level
5. Report

Step 1: Load or Create Test Plan

Resolve the test plan using these rules in order:

1. Explicit path — If a file path was passed, use it
2. Explicit slug — resolve to .turbo/test-plans/<slug>.md
3. Anchoring artifact — If the work under test is anchored to a plan, resolve to .turbo/test-plans/<that-slug>.md when that file exists
4. Single file — Glob .turbo/test-plans/*.md. If exactly one file exists, use it
5. Most recent — If multiple files exist, use the most recently modified
6. Legacy fallback — .turbo/test-plan.md if .turbo/test-plans/ does not exist
7. Nothing found — run the $create-test-plan skill first, then use the plan it writes

If multiple test plans exist and the most-recent choice is non-obvious, use request_user_input to let the user pick from the candidates.

Read the resolved test plan and state its path.

Unless an explicit path or slug was passed, confirm the resolved plan still describes the work under test:

• Unavailable branch state — a scenario's steps require a branch that no longer resolves in the repository
• Completed prior run — every checkbox is already ticked and no recorded result is FAIL or PARTIAL
• Superseded context — the plan's Context section names work that changes merged since the plan was written have reversed or removed

When a signal fires, output the signal and the scenarios it affects as text. For a superseded Context, name the scenarios that exercise the reversed or removed work. Then use request_user_input to offer:

• Regenerate — run the $create-test-plan skill with the resolved path, and use the plan it writes
• Execute anyway — the signal is a false positive
• Pick another plan — resolve to a different test plan file, then confirm that plan against these same signals

If the user specifies a narrower scope, filter the plan to relevant scenarios rather than executing all of them. Reserve filtering for that case: a superseded plan keeps scenarios that each look plausible alone, so trimming it preserves the wrong ones.

Step 2: Determine Testing Approach

Use the approach specified in the test plan. If the plan does not specify one, determine it using the same logic as $create-test-plan Step 2.

Step 3: Run $user-experience Skill (When User-Facing)

If the app has a user-facing surface (UI, screens, commands, messages, or any behavior a user sees or does), run the $user-experience skill to load the UX lens before executing tests, so usability concerns surface while interacting with the app. When it is unclear whether the surface is user-facing, use request_user_input to ask rather than skipping silently. Skip this step for test targets with no user-facing behavior (internal library or infrastructure).

Step 4: Execute Tests by Level

Work through each level sequentially. Complete all tests in a level before moving to the next.

Execution Loop (Per Test)

1. Set up the preconditions described in the test scenario
2. Perform the exact steps
3. Capture the result (screenshot, output, or state observation)
4. Compare against the expected outcome
5. Record PASS, FAIL, or PARTIAL with details
6. When the UX lens is loaded, note any usability observation it surfaces, kept separate from the verdict

When a scenario's preconditions need privileged state or a second participant (an entitlement or plan tier, an elevated role, seed data, a second concurrent client or session), provision it through a path the project already exposes for development and run the scenario. Record PARTIAL only after an attempt to provision failed, naming the precondition that could not be provisioned and what was tried.

When the scenario's output is consumed by another system, withhold PASS until that system accepts it. Decoding a token, reading a response body, or confirming a row exists shows only that the artifact was produced. Stand up the consumer under the same isolation and cleanup rules as any other service this run starts, and exercise its own flow. When standing it up is not possible, record PARTIAL and name which half is unproven. PARTIAL counts as not passed everywhere a verdict is tallied or gated.

Level Progression

1. Level 1: Basic Functionality — If any Level 1 test does not pass, report early and use request_user_input to ask whether to continue. Basic failures may indicate the feature is too broken for deeper testing.
2. Level 2: Complex Operations — Execute all tests regardless of individual failures.
3. Level 3: Adversarial Testing — Execute all tests. Failures here are expected and valuable.
4. Level 4: Cross-Cutting Scenarios — Execute all tests.

If a project-specific testing skill or MCP tool was identified in Step 2, use that. The paths below are fallbacks.

Web App Path

Reuse a running dev server only when this session started it. Otherwise start one on a port this run selected and wait for it to be ready. Confirm it bound to that port before sending it traffic — a failed bind leaves another agent's service answering. Move to another port when the port is taken; report the error and stop when the server itself failed to start. Use the browser-use@openai-bundled plugin to interact with the app.

UI/Native App Path

Launch the app. Use the computer-use@openai-bundled plugin to interact with the UI.

CLI Path

Run commands directly.

Step 5: Report

Present results organized by level:

Exploratory Test Results:

## Level 1: Basic Functionality (X/Y passed)
- [PASS] Test name: description
- [FAIL] Test name: description — [what went wrong]
- [PARTIAL] Test name: description — [which half is unproven]

## Level 2: Complex Operations (X/Y passed)
- [PASS] Test name: description
- [FAIL] Test name: description — [what went wrong]
- [PARTIAL] Test name: description — [which half is unproven]

## Level 3: Adversarial Testing (X/Y passed)
- [PASS] Test name: description
- [FAIL] Test name: description — [what went wrong]
- [PARTIAL] Test name: description — [which half is unproven]

## Level 4: Cross-Cutting Scenarios (X/Y passed)
- [PASS] Test name: description
- [FAIL] Test name: description — [what went wrong]
- [PARTIAL] Test name: description — [which half is unproven]

Overall: X/Y passed across all levels

Report usability observations from the UX lens below the level results, separately from the defects. A scenario can pass every functional check and still surface a usability concern.

## Usability Observations
- [UX] <observation> — names the UX context it touches (Understanding, Bridging, or Flowing) and the goal mismatch or friction it creates

For each failure, include the relevant screenshot, output, or state observation.

When the change under test spans several repositories, add a per-repo view of the findings below the usability observations, naming a suggested fix site for each.

Update the resolved test plan file by checking off completed tests and annotating results.

Then call update_plan to mark this step completed and continue with the next step of the active workflow.

Rules

• Always clean up: close only the browser sessions this run opened, by name, and stop the dev servers and other services this run started. Capture the PID of each server or service this run starts and stop it by that PID rather than by a name or command-line pattern, which also matches an identically named process a concurrent agent is running. Stop the process group rather than the captured PID alone — a server started behind a wrapper outlives its parent — and confirm the port released before reporting cleanup complete. Never close all browser sessions at once — concurrent agents may share the browser daemon, so a blanket close is cross-agent destruction.
• Treat a permission or scope granted mid-run to unblock a scenario as something this run created: before reporting cleanup complete, verify the production code never needs it, then ask for it to be revoked in the report. Name the call sites checked there too.
• Isolate shared process state so concurrent or sub-agent runs don't collide: bind dev servers and services to unique ports, scope tmux sessions (tmux -L <name>), give each browser session a unique name so cleanup can target only its own, and write screenshots and other scratch state to absolute paths under a unique scratch directory outside the repository under test. Derive each such identifier once and reuse that exact value in every later command, writing it as a literal or reading it back from a note under the run's scratch directory. A value recomputed per shell, such as $$, differs between the command that creates a resource and the command that releases it, so cleanup releases something it never created and reports success while the real resource leaks. A port picked as unique may already be held by a concurrent agent, so check it before binding and move to another when it is taken, leaving the incumbent running.
• Never modify application code. This skill is read-only verification. Report failures without attempting to fix them.
• If the dev server fails to start, report the error and stop.
• Tail app logs in a background shell for errors or warnings while running tests, so backend failures surface alongside test observations.
• To diagnose failures, run the $investigate skill on the test report.