agency-test-results-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external test result data, which creates a potential surface for indirect prompt injection if the ingested data contains malicious instructions targeting the LLM.
- Ingestion points: The
TestResultsAnalyzerclass inSKILL.mdusespd.read_json(test_results_path)to load data from an external file path. - Boundary markers: The provided instructions and code template do not define specific delimiters or instructions to the agent to ignore potentially embedded commands within the test data.
- Capability inventory: The skill's capabilities are limited to data analysis and reporting using libraries such as
pandasandscikit-learn; it does not exhibit high-risk capabilities like arbitrary shell command execution, network exfiltration, or persistence mechanisms. - Sanitization: The code template does not show explicit sanitization or validation logic for the content of the JSON files being processed.
Audit Metadata