pseo-metadata
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill guides the implementation of a metadata system that pulls from dynamic content sources, creating a pathway for untrusted data to enter the agent's context.
- Ingestion points: Untrusted content enters the workflow through the
getPageDatafunction and the resultingPageDataobject used inSKILL.md(Implementation Steps 1 and 8). - Boundary markers: The instructions and code templates lack explicit boundary markers or warnings to treat the content of the
PageDatafields as untrusted. - Capability inventory: The skill is configured to allow
Write,Edit,Bash,Read,Glob, andGreptools, which defines the potential impact area if the agent is influenced by injected instructions. - Sanitization: There is no evidence of sanitization, HTML escaping, or validation logic in the provided TypeScript examples to prevent malicious content in the data fields from affecting downstream consumers or the agent itself.
Audit Metadata