security-alert-monitor

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: Technical review of the skill's workflow and instructions found no evidence of malicious behavior, unauthorized data exfiltration, or attempts to bypass security guardrails.- [EXTERNAL_DOWNLOADS]: The skill references external resources for platform configuration and user documentation.
  • Evidence: References an MCP endpoint at mcp.igpt.ai and documentation on raw.githubusercontent.com.- [PROMPT_INJECTION]: The skill's primary function involves processing untrusted external data (email content), which creates a potential surface for indirect prompt injection.
  • Ingestion points: Email content is retrieved using the search tool and passed to the agent for analysis.
  • Boundary markers: The analysis prompt does not employ specific delimiters or ignore instructions for the email body content.
  • Capability inventory: The skill is restricted to information retrieval and structured summarization; it has no access to tools for file system modification or outbound network requests.
  • Sanitization: No content filtering or validation is performed on the ingested email data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 01:00 AM
Security Audit — agent-trust-hub — security-alert-monitor