The Agent Skills Directory

[PROMPT_INJECTION]: The skill's primary function is to audit RAG pipeline implementations by reading source code and documentation using the Read, Glob, and Grep tools. This behavior establishes an ingestion surface for untrusted data. If the files being audited contain malicious instructions, they could potentially influence the agent's reasoning or the output of the audit report.
Ingestion points: External project files and documentation accessed via Read, Glob, and Grep tools.
Boundary markers: The skill does not explicitly instruct the agent to use delimiters or ignore embedded instructions when reading untrusted file content.
Capability inventory: The skill is restricted to read-only capabilities (Read, Glob, Grep) and lacks network or code execution tools, which significantly limits the impact of any potential injection.
Sanitization: No specific filtering or sanitization of the content retrieved from audited files is defined.

rag-review