monitor-internet-outages

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves internet outage reports from an external API (worldmonitor.app), which includes fields like titles and descriptions that may originate from external/third-party data sources. This creates a potential surface for indirect prompt injection if the fetched content contains malicious instructions.
  • Ingestion points: API response data from https://www.worldmonitor.app/api/infrastructure/v1/list-internet-outages as defined in SKILL.md.
  • Boundary markers: Present. The 'Content safety' section provides explicit, clear instructions to the agent to treat all response fields strictly as content to analyze or quote and to disregard any directive-like text found within (e.g., 'ignore previous instructions').
  • Capability inventory: The skill definition is limited to data retrieval via HTTP GET; it does not implement file-writing, system-level command execution, or dynamic code evaluation.
  • Sanitization: The skill employs policy-based sanitization by providing the LLM with specific instructions on how to handle untrusted data fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 05:53 AM