analyze-dashboard

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a legitimate analysis assistant, using platform-specific tools to retrieve metrics and structure. It does not perform any high-risk operations such as arbitrary code execution, sensitive file access, or unauthorized network communication.- [PROMPT_INJECTION]: The skill ingests untrusted user feedback from Amplitude:get_feedback_insights and Amplitude:get_feedback_mentions. Evidence chain: 1. Ingestion points: User feedback tools in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Limited to data retrieval and summarization; no dangerous tools or subprocess calls. 4. Sanitization: Absent. This is a potential indirect prompt injection surface but is considered safe given the limited capabilities and intended use-case.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 08:19 PM
Security Audit — agent-trust-hub — analyze-dashboard