skills/cogni-work/insight-wave/ask/Gen Agent Trust Hub

ask

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill mentions a reference link to a GitHub repository belonging to the vendor (cogni-work). This is a static documentation link and does not represent a runtime download of executable code.
  • [COMMAND_EXECUTION]: The skill uses grep and bash via the allowed-tools configuration to search local documentation files. These operations are scoped to the plugin's root directory and are used for reading the bundled wiki index and pages.
  • [PROMPT_INJECTION]: There are no patterns of prompt injection detected. The instructions prioritize grounded answers from the local wiki and explicitly forbid the model from answering from its own memory or overriding safety guidelines.
  • [DATA_EXFILTRATION]: No network operations or unauthorized file access patterns were detected. The skill reads local markdown files from a predefined path for information retrieval.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 01:50 PM
Security Audit — agent-trust-hub — ask