performance-analysis

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes performance data from external campaign management tools, which represents a potential indirect prompt injection surface if campaign names or metadata were manipulated.
  • Ingestion points: Performance figures and outliers are read via metrics.account-overview, metrics.sequence, and sender.health tools as described in SKILL.md.
  • Boundary markers: The instructions do not define explicit delimiters or 'ignore' warnings for the data being analyzed.
  • Capability inventory: The skill's primary capabilities are analytical; it includes one confirmed action (pausing campaigns) via the campaign-launch skill, but lacks network or file-system write access.
  • Sanitization: No specific sanitization or escaping of the metrics data is mentioned in the prompt logic.
  • [SAFE]: The skill enforces a human-in-the-loop requirement, stating it is 'Read-only by default' and that actions like pausing campaigns only occur on 'explicit approval'.
  • [SAFE]: Tool usage is limited to scoped diagnostic and measurement operations, adhering to the principle of least privilege for its stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 06:47 PM
Security Audit — agent-trust-hub — performance-analysis