The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to ingest and process untrusted data from the user's codebase.
Ingestion points: Ingests content from source code files, .env files, and configuration files via the Read and Grep tools.
Boundary markers: The instructions do not specify the use of delimiters or boundary markers when interpolating file content into the agent context.
Capability inventory: The skill is restricted to Read and Grep tools and does not have network access or arbitrary code execution capabilities.
Sanitization: No explicit sanitization or filtering logic is defined for the ingested data.
Risk assessment: Although the ingestion of untrusted data is a known vector, the risk is considered safe given the skill's primary purpose is security auditing and its lack of high-privilege capabilities.

secret-scanner