The Agent Skills Directory

[SAFE]: The skill focuses on transforming product requirements into test documentation. It does not contain instructions to bypass safety guidelines or override system prompts.
[DATA_EXFILTRATION]: The skill operates on local files provided by the user and writes its output to a specific tests/ directory. There are no network calls or attempts to send sensitive data to external domains.
[COMMAND_EXECUTION]: While the skill uses file system tools (Read and Write), it follows a well-defined workflow and includes explicit instructions in Step 6 to use a 'sanitized version' of titles for filenames, which serves as a mitigation against path traversal risks.
[INDIRECT_PROMPT_INJECTION]: The skill has a standard attack surface for indirect injection as it processes untrusted PRD documents:
Ingestion points: Uses the Read tool to ingest requirement documents in Step 1.
Boundary markers: No specific delimiters or 'ignore' instructions are provided for the content of the read files.
Capability inventory: Access to Read and Write tools and the ability to query the user.
Sanitization: Explicitly instructs the agent to sanitize filenames before writing.

test-cases