offensive-network-attacks
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external content, including source code, binaries, and network configurations, to identify vulnerabilities. This creates a surface for indirect prompt injection where malicious instructions embedded in the audited files could influence the agent's behavior.
- Ingestion points: Identified in
SKILL.mdunder the Discovery and Surface Inventory phases (Source search / REA binary & bundle inspection). - Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the audited data.
- Capability inventory:
SKILL.mddescribes capabilities including static and binary analysis, implementation of remediation code, and execution of automated regression tests (npm test). - Sanitization: Absent; the skill does not mention sanitizing, escaping, or validating the untrusted content before processing or interpolating it into the agent's context.
Audit Metadata