The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash, Glob, and Grep tools to search the workspace for configuration files and specific rule patterns. These operations are restricted to scanning and reporting on local files as described in the workflow.
[PROMPT_INJECTION]: The skill possesses an ingestion surface for indirect prompt injection because it reads and processes untrusted content from the workspace (e.g., CLAUDE.md, CODEX.md, .cursorrules).
Ingestion points: Reads various markdown and JSON configuration files across the entire workspace.
Boundary markers: The instructions do not specify the use of delimiters or ignore-instructions for the processed data.
Capability inventory: The skill has access to Write, Edit, and Bash tools, which could theoretically be manipulated by malicious content within the audited files.
Sanitization: No explicit sanitization or validation of the workspace content is mentioned before it is used to generate reports or apply automated fixes.
[SAFE]: The skill's behavior aligns with its stated purpose of workspace maintenance. It identifies hardcoded absolute paths and stale 'Last Updated' dates to improve portability and reliability of agent configurations.

agent-config-audit