owasp
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows established patterns for multi-agent orchestration without introducing malicious instructions or dangerous capabilities beyond its intended security analysis function.
- [PROMPT_INJECTION]: The skill instructions include clear structural markers for subagents, such as "STEP 1", "STEP 2", and "IMPORTANT", which serve to guide the agent's behavior for its primary purpose rather than attempting to bypass safety protocols.
- [INDIRECT_PROMPT_INJECTION]: As a security analysis tool, this skill processes untrusted code files which represents an attack surface for indirect prompt injection.
- Ingestion points: Code files resolved from the
--scopeflag are read and processed for relevance using pattern matching and passed to subagents for analysis. - Boundary markers: The dispatcher uses clear section headers (e.g.,
FILES:,STEP 1:) in its subagent prompt templates to help isolate data from instructions. - Capability inventory: The skill utilizes file system search (Glob/Grep) and subagent dispatching (Task tool) to perform its duties.
- Sanitization: The dispatcher performs logic-based filtering of categories based on file contents but does not explicitly sanitize the raw file data before passing it into the subagent context windows.
Audit Metadata