keyword-expansion
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No attempts to override agent behavior, bypass safety filters, or extract system prompts were detected.
- [Data Exposure & Exfiltration] (SAFE): The skill only interacts with non-sensitive local markdown files (queries.md, DECISIONS.md). No network calls or sensitive path accesses are present.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill contains no scripts, package dependencies, or dynamic code execution logic.
- [Indirect Prompt Injection] (LOW):
- Ingestion points: Reads from
queries.mdandDECISIONS.md. - Boundary markers: Not explicitly defined in the prompt instructions.
- Capability inventory: Restricted to reading and writing text to markdown files. No subprocess or network capabilities.
- Sanitization: Not present, but risks are minimal as the output is constrained to keyword lists.
Audit Metadata