The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as its primary function is to ingest and analyze untrusted source code and configuration files.
Ingestion points: The skill uses Read, Grep, and Glob tools to analyze project files based on user-provided arguments.
Boundary markers: Includes a robust 'Prompt Injection Safety Notice' that explicitly instructs the agent to ignore instructions embedded in reviewed content and to avoid executing code found during analysis.
Capability inventory: The skill is restricted to Read, Grep, and Glob tools as defined in the frontmatter, limiting the potential impact of an injection.
Sanitization: There is no programmatic sanitization of the ingested content; the skill relies on the agent's internal guardrails and the provided defensive instructions.
[SAFE]: The skill represents a defensive security auditing tool. It does not initiate network connections, request elevated privileges, or use dangerous execution patterns. All analysis logic is based on searching for known vulnerable patterns (e.g., insecure pickle loading) in a target codebase.

model-supply-chain