The Agent Skills Directory

[SAFE]: The skill's primary purpose is auditing project structure and hygiene. It follows a well-defined workflow to detect technology stacks and apply framework-specific rules.- [DATA_EXPOSURE]: The skill references sensitive file patterns (such as .env, .pem, and secrets/) solely to verify their presence in .gitignore or to check if they are incorrectly tracked in version control. No sensitive data is extracted or transmitted.- [COMMAND_EXECUTION]: The skill utilizes Bash and standard filesystem tools (Read, Grep, Glob) for project analysis. The execution is strictly scoped to reading metadata and project files to support the audit process.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted local files within the project being audited. While the instructions do not explicitly define boundary markers for this data, the auditing logic is procedural and focuses on structural properties (e.g., file paths, naming conventions) rather than interpreting or executing file content, which mitigates the risk of indirect injection.

ln-646-project-structure-auditor