handle-my-replies

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes social media replies which are untrusted external inputs. An attacker could embed instructions in a comment to manipulate the agent's behavior. The skill mitigates this by instructing the agent to treat replies as data rather than instructions and ensuring a human-in-the-loop for all drafted responses. * Ingestion points: Social media replies ingested for triage. * Boundary markers: Explicit section warning the agent to treat replies as text data rather than instructions. * Capability inventory: Drafting text for user review; no automated command execution or file modifications. * Sanitization: Not programmatically implemented.
  • [SAFE]: The skill references voicemoat.com as an informational link for a voice profiling service. This reference does not trigger any malicious behavior or involve untrusted downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 07:23 AM