google-cloud-waf-performance-optimization
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to inspect repository files and incorporate data from external SandBase tool calls into its context, creating a potential attack surface for indirect instructions.
- Ingestion points: The skill ingests untrusted data from existing codebase files (SKILL.md, workflow step 1) and external documentation or model outputs retrieved via the SandBase API (SKILL.md, SandBase boundary section).
- Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or triple backticks) or explicit "ignore embedded instructions" warnings when processing these external or local inputs.
- Capability inventory: The skill possesses the ability to execute platform tools (
sandbase_run), write implementations, and review configurations, which could be influenced by malicious content in the processed data. - Sanitization: There is no evidence of sanitization, escaping, or filtering of the content retrieved from the repository or external sources before it is interpolated into the agent's context.
Audit Metadata