no-polling-agents

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The skill mentions reading a local status file (".claude/cache/status.txt"). This is a standard path for agent state and does not involve sensitive user data or external exfiltration.
  • [Unverifiable Dependencies] (SAFE): No external libraries or remote code sources are utilized.
  • [Prompt Injection] (SAFE): There are no attempts to override agent behavior or bypass safety constraints.
  • [Indirect Prompt Injection] (LOW): The skill advises reading from a file populated by other agents. While this creates a potential surface for indirect injection if an agent reads malicious status data, it is a standard operational pattern and the skill itself is safe.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:03 PM