test-reporting-analytics

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from test results and quality metrics to generate reports and dashboards. While no high-privilege capabilities like shell access are exposed, the generated content could influence downstream agents in a multi-agent system. \n- Ingestion points: Inputs such as testResults, historicalData, and failures in evals/test-reporting-analytics.yaml. \n- Boundary markers: No delimiters or instructions to ignore embedded commands are defined in the skill logic. \n- Capability inventory: The skill uses Task() and FleetManager to coordinate with other agents. No direct file or network operations are present. \n- Sanitization: There is no evidence of validation or sanitization of input data before its inclusion in reports.\n- [SAFE]: No hardcoded credentials, malicious obfuscation, or unauthorized remote code execution patterns were found. The use of standard utilities like jq in the evaluation configuration is consistent with normal development practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 12:34 AM