08-business-process-modeling
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external project files which serves as a potential vector for indirect prompt injection. Malicious instructions embedded in elicitation logs or stakeholder registers could attempt to influence the agent's modeling logic or output.
- Ingestion points: Files located at
projects/<ProjectName>/<phase>/<document>/elicitation_log.md,stakeholder_register.md,features.md, andbusiness_rules.mdare defined as required or optional inputs in the Quick Reference section ofSKILL.md. - Boundary markers: The workflow lacks explicit delimiters or instructions to the agent to ignore or isolate instructions found within the input text files.
- Capability inventory: The skill possesses file-write capabilities, specifically the ability to create
business_process_models.mdin Step 7. - Sanitization: There are no documented procedures for sanitizing, escaping, or validating the content extracted from the source documents before processing.
Audit Metadata