The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill analyzes external code files, which creates an attack surface for indirect prompt injection.
Ingestion points: Content is read from local files using the Read, Grep, and Glob tools.
Boundary markers: The instructions do not include specific boundary markers or warnings to the agent to treat ingested file content strictly as data rather than instructions.
Capability inventory: The agent is provided with the Bash tool, which represents a high-privilege capability that could be misused if the agent is influenced by malicious instructions embedded in the reviewed code.
Sanitization: The skill lacks procedures for sanitizing, escaping, or validating the content of the files before the agent processes them.

maintainability-reviewer