polylane-automations
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a framework where agents execute instructions based on external event triggers, including webhooks, alerts, and GitHub events. This creates a surface where malicious instructions embedded in those external payloads could influence agent behavior.
- Ingestion points: External data enters the agent context via the
--triggerpayloads (alert, webhook, github.*) as described in SKILL.md. - Boundary markers: The provided templates and instructions do not include the use of delimiters or warnings to ignore instructions embedded within the trigger data.
- Capability inventory: The system possesses significant capabilities, including performing side-effects like opening issues, merging pull requests, and rolling back deployments.
- Sanitization: The authoring checklist lacks guidance on validating or sanitizing external content before it is processed by the agent instructions.
Audit Metadata