clari-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external revenue data exports which serves as a potential vector for indirect prompt injection if those exports contain malicious content.\n
- Ingestion points: Untrusted data is processed through the
forecasts,actuals,entries,current, andpreviouslist parameters in the Python scripts defined inSKILL.md.\n - Boundary markers: The instructions lack specific delimiters or boundary markers to ensure the agent ignores instructions that may be embedded within the data.\n
- Capability inventory: The skill uses
Bash(python3:*)to execute the data analysis logic.\n - Sanitization: The logic processes data assuming a valid schema but does not implement content sanitization or validation to filter out potential instructions within data fields.
Audit Metadata