junit-xml-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process JUnit XML reports, which are often generated by external, potentially untrusted sources.
- Ingestion points: XML report files (e.g.,
reports/*.xml) are parsed to extract test metadata, failure messages, and stack traces. - Boundary markers: The provided parsing logic does not implement delimiters or 'ignore' instructions to prevent the model from obeying instructions potentially embedded within test failure messages or class names.
- Capability inventory: The results of the analysis are written to the file system as JSON data and Markdown reports (e.g.,
slowest.json,flake-report.md). - Sanitization: There is no evidence of sanitization or escaping of the content extracted from the XML files before it is included in the generated reports.
- [NO_CODE]: The skill does not include standalone executable script files. All processing logic is provided as instructional code snippets within Markdown files or as part of the evaluation task descriptions.
Audit Metadata