halo-cli-moderation-notifications

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the 'halo' command-line interface to perform moderation actions such as listing, approving, and deleting comments and notifications. This is appropriate for its stated purpose as a content management tool.
  • [SAFE]: No evidence of prompt injection, data exfiltration, or obfuscated code was found in the provided skill file.
  • [SAFE]: Indirect Prompt Injection Surface Analysis:
  • Ingestion points: Reads content from comments and notifications via 'halo comment list' and 'halo notification list'.
  • Boundary markers: None present.
  • Capability inventory: Includes 'halo comment delete' and 'halo comment create-reply' which allow modifications to the environment.
  • Sanitization: None present.
  • Note: This risk is standard for moderation-focused skills and is mitigated by the agent's general safety guardrails and the use of the official CLI.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 02:34 AM