The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run standard development verification tools including ruff check . and pytest. These are legitimate operations within the context of a code maintenance and refactoring workflow.
[INDIRECT_PROMPT_INJECTION]: The skill operates by scanning and modifying local source code, which constitutes an attack surface for indirect prompt injection. A malicious codebase could potentially contain comments or strings designed to misdirect the agent during the refactoring process. However, this is an inherent risk of the skill's primary function (code refactoring) and no specific vulnerabilities or unsafe data handling practices were detected.
Ingestion points: Reads project source code during the 'Scan the Codebase' phase (SKILL.md).
Boundary markers: None explicitly defined in the instructions.
Capability inventory: File modification (fixing code) and command execution (ruff, pytest) (SKILL.md).
Sanitization: None specified for handling untrusted data in processed files.

techdebt