search-strategy
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is composed entirely of instructional text and tables. It does not contain any executable scripts, binary files, or network-enabled commands.
- [NO_CODE]: Analysis confirms the absence of functional code in the skill; it operates as a set of logical instructions for the AI agent to follow during search tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for transforming user input into tool queries, which is a standard surface for indirect prompt injection. 1. Ingestion points: Natural language queries processed in the 'Query Decomposition' section of SKILL.md. 2. Boundary markers: Absent from the transformation examples. 3. Capability inventory: Calls to chat, knowledge base, and project tracking tools as described in the 'Source-Specific Query Translation' section. 4. Sanitization: Not specified in the instructions; implementation relies on the agent's base safety protocols.
Audit Metadata