tradecraft-complex-engagements
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted data from targets (logs, source code, headers) which creates a potential surface for indirect prompt injection.
- Ingestion points: Target logs, application source code, network headers, and filenames (SKILL.md).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between target data and its own task instructions.
- Capability inventory: The methodology assumes the agent has access to file system operations for record-keeping and potentially network tools for reconnaissance.
- Sanitization: No mentions of escaping or sanitizing data harvested from target environments.- [SAFE]: The skill serves as a purely instructional framework for professional security assessments. It explicitly references Rules of Engagement (ROE) and advises caution, such as reviewing proof-of-concept code before execution. No code, obfuscation, or persistence mechanisms were detected.
Audit Metadata