The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, potentially untrusted source code provided via $ARGUMENTS.
Ingestion points: The skill uses the Read and Grep tools to load the content of files or directories into the agent's context for scanning (SKILL.md).
Boundary markers: The instructions do not define explicit delimiters or include specific guidance for the agent to ignore or isolate natural language instructions that might be present within comments or strings in the scanned code.
Capability inventory: The skill utilizes the Read and Grep tools to access and process filesystem data (SKILL.md).
Sanitization: There is no evidence of content sanitization or validation to filter out potential prompt injection attacks embedded in the source data before the agent processes it.

red-flags