The Agent Skills Directory

[SAFE]: The skill performs legitimate static analysis tasks to enforce architectural boundaries and best practices. No evidence of prompt injection, data exfiltration, or persistence mechanisms was found.
[COMMAND_EXECUTION]: The skill uses allowed system tools including Bash, Grep, and Read to scan the codebase and generate audit reports. These operations are scoped to the project root and the designated output directory, consistent with the worker's auditing purpose.
[DATA_EXPOSURE]: The skill includes checks for direct environment variable access (e.g., os.environ) to ensure they are handled in configuration layers rather than domain logic. This is an architectural best-practice check and does not involve the theft of secrets.

ln-642-layer-ownership-boundary-auditor