skills/peterbamuhigire/chwezi-sdlc-documentation/13-saas-billing-and-metering-spec/Gen Agent Trust Hub
13-saas-billing-and-metering-spec
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (project documentation) which could contain malicious instructions aimed at overriding agent behavior or poisoning the resulting specification.
- Ingestion points: The
logic.promptinstructions and Step 1 of the Core Instructions require the agent to read various local files includingPRD.md,vision.md,pricing & packaging spec,Multi_Tenancy_Architecture_Spec.md,HLD.md, andtech_stack.md. - Boundary markers: The instructions do not define clear delimiters or provide the agent with a 'security context' to ignore instructions embedded within the ingested project files.
- Capability inventory: The skill is primarily instructional and focused on documentation; it allows reading files and writing a new markdown file (
Billing_And_Metering_Spec.md). It does not demonstrate network exfiltration, arbitrary command execution, or dynamic code evaluation capabilities. - Sanitization: There is no evidence of content filtering, escaping, or schema validation for the external project data being processed.
Audit Metadata