confluence-expert
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is primarily instructional and contains well-structured templates. The included scripts for content auditing, label analysis, and space structure evaluation use standard libraries and do not engage in risky network or file system behaviors.
- [PROMPT_INJECTION]: The skill includes an inherent surface for indirect prompt injection. 1. Ingestion points: The scripts scripts/content_health_auditor.py, scripts/label_taxonomy_analyzer.py, and scripts/space_structure_analyzer.py read documentation metadata (titles, labels, etc.) from JSON files. 2. Boundary markers: The processing logic does not implement markers to distinguish data from potential embedded instructions. 3. Capability inventory: The scripts are limited to data analysis and reporting; they possess no capabilities for network exfiltration or dangerous command execution. 4. Sanitization: There is no sanitization of the input metadata before it is processed and summarized in the health reports.
Audit Metadata