org-health-diagnostic
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted natural language data provided by the user, creating a surface where malicious instructions could be embedded within organizational metrics.
- Ingestion points:
SKILL.mdallows users to 'describe your metrics in natural language' or 'paste your key metrics'. - Boundary markers: There are no instructions or delimiters defined to isolate user-provided data from agent instructions.
- Capability inventory: The skill is configured to execute a local script
scripts/health_scorer.py(Category: command_execution). Analysis of the script confirms it only performs local calculations and console output; it does not possess network or persistent file-writing capabilities. - Sanitization: No sanitization or validation of the input metrics is implemented in the provided prompt or script logic.
- [COMMAND_EXECUTION]: The skill uses a local script
scripts/health_scorer.pyas defined in theSKILL.mdfrontmatter and the documentation for processing diagnostic data.
Audit Metadata