data-client-manager
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and act upon data from external event streams, which introduces a vulnerability surface where malicious instructions could be embedded in the data.
- Ingestion points: The documentation in
SKILL.mddescribes a "Websocket/SSE push streams" use case where external messages triggercontroller.set()to update the application state. - Boundary markers: The instructions do not specify or require the use of boundary markers or instructions to ignore embedded content when processing these external streams.
- Capability inventory: The skill utilizes a
Controllerthat provides significant capabilities, includingcontroller.set()for state modification andcontroller.resetEntireStore()for clearing data. - Sanitization: The provided code templates for
TimeManagerandLoggingManagerdemonstrate direct processing of time or action payloads without explicit validation or sanitization of the content before it enters the store.
Audit Metadata