The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and analyze external files within a user-provided project path, which exposes the agent to untrusted data that could contain malicious instructions (Indirect Prompt Injection).
Ingestion points: The agent reads content from files at the provided <project_path> using tools like Read, Glob, and Grep as part of its 'Scanning Methodology'.
Boundary markers: The skill definition does not provide clear delimiters or 'ignore' instructions to help the agent distinguish between data to be scanned and potential adversarial instructions embedded within the scanned content.
Capability inventory: The skill authorizes the use of powerful tools including Bash, Read, Glob, and Grep, which could be exploited if the agent inadvertently follows instructions found in scanned files.
Sanitization: There are no explicit instructions for the agent to sanitize, escape, or validate the content read from the external project directory before processing it.

vulnerability-scanner