search-skill

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions to override agent behavior or bypass safety filters were found in the instructions or code.
  • [Data Exposure & Exfiltration] (SAFE): The code interacts exclusively with a local database object passed at runtime and does not perform network operations or access sensitive system files like SSH keys or credentials.
  • [Remote Code Execution] (SAFE): There are no patterns involving remote script fetching, package installations, or dynamic code evaluation (e.g., eval, exec, or subprocess calls).
  • [Indirect Prompt Injection] (SAFE): The skill processes user-controlled search terms (keywords, tags). However, the evidence chain shows no dangerous capabilities (network, write access, or execution) that could be triggered by malicious strings in the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 12:23 PM