skills/peterbamuhigire/business-plan-skills/meta-agent-revenue-recognition-policy/Gen Agent Trust Hub
meta-agent-revenue-recognition-policy
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as executed contracts, pricing primitives, and financial records as required inputs.
- Ingestion points: Data enters the agent context through client-supplied project files, source registers, and accounting records as defined in the 'Required Inputs' table in SKILL.md.
- Boundary markers: The skill provides 'Decision, stop, and recovery controls' to halt if evidence is missing, but lacks explicit instructional delimiters to isolate processed text from the agent's core instructions.
- Capability inventory: The skill is restricted by its 'Capability and Permission Boundaries' to read-only operations and document drafting; it has no permissions for network writes, fund commitment, or system modification.
- Sanitization: There is no evidence of automated sanitization or filtering for the content of the ingested financial documents before processing.
Audit Metadata