profile-game-performance
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data sources which represents a potential injection vector.
- Ingestion points: The skill processes telemetry from device farms, player crash reports, and prior profiling sessions as context (referenced in the Security Guardrails section).
- Boundary markers: The instructions do not define specific structural delimiters (e.g., XML tags) to isolate untrusted telemetry data from the agent's instructions.
- Capability inventory: The skill has access to a wide range of sensitive tools including
execute_command,write_file,edit_file, andread_fileacross the project environment. - Sanitization: The skill includes a conceptual mitigation to treat captured metrics as untrusted data, but it lacks specific programmatic sanitization or validation steps for the content of these reports.
Audit Metadata