session-security-testing
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (logs, configuration files, and security scan evidence) as part of its core function, creating a surface for indirect prompt injection.\n
- Ingestion points: The skill ingests data from user-provided prompts and static fixtures (as noted in prompts/session-security-testing.md).\n
- Boundary markers: The skill utilizes an "input audit" to separate known facts from assumptions, though this is a procedural instruction for the LLM rather than a technical isolation mechanism.\n
- Capability inventory: The skill is restricted to text output and does not possess capabilities for file writing, network access, or subprocess execution in its provided configuration.\n
- Sanitization: No specific sanitization or escaping logic is implemented for the ingested data before it is included in the model's processing context.
Audit Metadata