ict-services-firm-business-plan
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a conceptual framework for business analysis and does not contain executable scripts, malicious command strings, or unauthorized network operations. References point to localized markdown components within the skill environment.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for processing sensitive external business data, including CRM logs, payroll records, and financial documents. This creates a theoretical ingestion surface for adversarial content. However, the risk is mitigated by the inclusion of a 'Capability Contract' that defines clear boundaries, restricting the agent to read-only operations and prohibiting unauthorized system modifications. Evidence chain: 1. Ingestion points: Service catalogs, CRM probabilities, and financial records (SKILL.md). 2. Boundary markers: The 'Capability Contract' and 'Decision Rules' provide instructional constraints. 3. Capability inventory: The skill is limited to data mapping, economic modeling, and plan generation. 4. Sanitization: No programmatic sanitization is defined, relying instead on structural prompts to guide agent behavior.
Audit Metadata