The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the find command via the Bash tool to search for generic directory names such as 'utils', 'helpers', or 'common' within the target project directory.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface because it reads and processes data from external files that could contain malicious instructions.
Ingestion points: Project source code, architecture documentation (CLAUDE.md), and configuration files (package.json, pyproject.toml) are read from the file system.
Boundary markers: The skill does not use delimiters or specific instructions to ensure the agent ignores potential commands embedded in the project files.
Capability inventory: The skill has access to Bash, Read, Glob, and Grep tools, allowing it to inspect and report on project status.
Sanitization: There is no evidence of content sanitization or validation performed on the ingested project data.
[SAFE]: The skill contains hardcoded absolute Windows paths (e.g., C:/Users/gurusharan.gupta/Agents/...) which disclose a local username. This is a best-practice violation for privacy and portability but does not represent an active security threat.

audit