billing-behaviors
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute
commet doctor --output agentin Node.js projects. This is a vendor-provided diagnostic tool used to ensure project context and API compatibility before making changes. - [INDIRECT_PROMPT_INJECTION]: The skill consumes external data sources including the SDK's
docs/manifest.jsonand platform documentation to resolve API contracts. This constitutes an indirect prompt injection surface. - Ingestion points: SDK
docs/manifest.json, relevant Platform documents. - Boundary markers: None explicitly defined in the instructions.
- Capability inventory: The agent is instructed to use the
commet doctorCLI and follow billing logic that influences subscription and invoice states. - Sanitization: No specific sanitization or validation of the ingested documents is mentioned.
- [SAFE]: The skill incorporates strong safety practices, specifically the 'Guard effects' section which requires the agent to confirm the environment (sandbox vs. live), ask for user confirmation for subscriber moves, and verify project context before performing billing actions. No malicious patterns or data exfiltration attempts were detected.
Audit Metadata