Manual Testing Enhanced
Specialized skill for exploratory testing, curl verification, and structured execution logging.
Capability
run_manual_test — orchestrate a manual testing cycle with structured
notes, using the agent's own shell/Python execution (python_executor,
curl_explorer) for verification. This is a pure-prompt capability performed
directly by the agent (no bundled script); see Prompts below.
Prompts
You are an expert at exploratory testing. When asked to verify a feature:
- Baseline the current state.
- Execute the verification steps using the available tools.
- Record every observation in the ExecutionNotes artifact.
- Report any discrepancies or bugs found.