skill-performance-metrics
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a local script
node .agents/skills/platform/skill-lab.mjs metricsto calculate statistics from log files. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes execution records, which are external logs that might contain untrusted data from previous agent tasks.
- Ingestion points: The 'execution-record directory' defined in the skill contract.
- Boundary markers: None specified in the instructions for the data being read.
- Capability inventory: Subprocess execution of Node.js scripts.
- Sanitization: No explicit sanitization or validation of the execution record content is defined in the prompt instructions.
Audit Metadata