test-reporting-analytics
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, such as test results, failure logs, pull request diffs, and historical metrics, to generate dashboards and predictive insights. This creates a standard attack surface for indirect prompt injection.
- Ingestion points: External data enters the agent's context through task parameters (e.g.,
codeChanges: prDiff,testResults,failures) as described inSKILL.mdand the evaluation test cases inevals/test-reporting-analytics.yaml. - Boundary markers: The provided report templates in
SKILL.mddo not include explicit boundary markers or instructions to ignore embedded commands within the ingested data. - Capability inventory: The skill's capabilities are limited to metrics aggregation, report generation, and trend analysis using tools like
jq. It does not perform sensitive file operations, network exfiltration, or arbitrary command execution. - Sanitization: While explicit sanitization logic is not provided in the instructions, the skill enforces strict structure through a JSON output schema (
schemas/output.json) which mitigates the impact of potential injection by requiring data to conform to specific types and patterns. - [SAFE]: No other security issues, such as hardcoded credentials, obfuscation, or malicious persistence mechanisms, were detected in the analyzed files.
Audit Metadata