hyperframes-registry
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 ingest and process data from external sources and the SandBase API, creating a surface for indirect prompt injection if the retrieved content contains malicious instructions.
- Ingestion points: External data sources and generative outputs retrieved through the
sandbase_runtool as described inSKILL.mdandreferences/sandbase-api-map.md. - Boundary markers: Instructions mention preserving "authorization boundaries" but do not provide specific delimiters or clear instructions to the agent to ignore commands within external data.
- Capability inventory: Includes tool access for remote capability discovery (
sandbase_discover), inspection (sandbase_inspect), and execution (sandbase_run). - Sanitization: The workflow does not explicitly define sanitization or filtering steps for content returned from external API calls.
Audit Metadata