The Agent Skills Directory

[DATA_EXPOSURE]: The skill is designed to read and analyze local source code and test files to provide a hygiene audit. While this involves accessing the local filesystem, the instructions are strictly focused on analysis and reporting within the agent's context. No network exfiltration or credential harvesting patterns were found.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data (user-provided source code and test files).
Ingestion points: Reads project files including README.md, documentation, test configuration, and production code (SKILL.md, Step 1).
Boundary markers: The instructions do not specify any delimiters or safety markers for the code being analyzed.
Capability inventory: The skill's capabilities are limited to file system read access and generating markdown reports. No file write or arbitrary shell execution capabilities are requested.
Sanitization: There are no explicit instructions to sanitize or escape content found within the analyzed files.
Risk Assessment: The risk is minimal as the skill's output is purely informational (a cleanup plan) and does not trigger secondary actions based on the content of the files.
[COMMAND_EXECUTION]: The skill mentions detecting subprocess or command usage in code, but these are listed as search patterns (signals) to identify 'dirty' tests in the user's codebase, not instructions for the agent to execute arbitrary shell commands.

unit-test-hygiene