pagerduty-automation

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill identifies and references an external MCP server configuration at https://rube.app/mcp. This is the intended endpoint for the Rube/Composio toolkit used for PagerDuty automation and is documented neutrally as it represents the skill's primary functional dependency.
  • [PROMPT_INJECTION]: An indirect prompt injection surface is present by design, as the skill processes untrusted external data from PagerDuty incidents.
  • Ingestion points: Untrusted data enters the context through tools such as PAGERDUTY_POST_INCIDENT_NOTE_USING_ID (incident notes), PAGERDUTY_GET_INCIDENT_ALERT_DETAILS (alert descriptions), and PAGERDUTY_CREATE_INCIDENT_RECORD (incident titles).
  • Boundary markers: No specific boundary markers or 'ignore embedded instructions' warnings are defined in the skill logic for handling external PagerDuty data.
  • Capability inventory: The skill possesses extensive management capabilities, including the ability to create and update services, escalation policies, on-call schedules, and teams.
  • Sanitization: No explicit sanitization, validation, or escaping of ingested strings is defined in the documentation or tool sequence logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 07:56 AM