org-health-diagnostic
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a framework for organizational health diagnostics using local scripts and standard instructions. No malicious patterns, such as credential theft, persistence, or remote code execution, were identified during the analysis.
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
scripts/org_health_scorer.py,scripts/span_of_control_analyzer.py, andscripts/engagement_benchmarker.py) to process user metrics. These executions are legitimate parts of the skill's functionality and do not involve unauthorized system commands or network connectivity. - [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from CSV files and user-supplied metrics, creating a potential surface for indirect prompt injection. However, the analysis is limited to data processing and structured reporting, which carries a low exploitation risk. Evidence chain: 1. Ingestion points:
scripts/span_of_control_analyzer.pyreads data from a user-provided CSV file path. 2. Boundary markers: No explicit markers or warnings to ignore embedded instructions are present in the scripts or instructions. 3. Capability inventory: The skill utilizes local file reading and script execution to generate reports; it lacks network access or elevated privileges. 4. Sanitization: The scripts perform basic string parsing and type casting but do not implement formal sanitization against injection patterns.
Audit Metadata