agent-memory-coordinator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and stores user-provided information for long-term persistence, which represents an indirect prompt injection surface.
- Ingestion points: Data provided during 'Store', 'Sync', and 'Namespace Management' operations as described in SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from executing commands embedded within stored memories.
- Capability inventory: The skill coordinates memory namespaces and synchronization but does not execute external scripts or privileged system commands within the provided configuration.
- Sanitization: There is no evidence of content filtering, escaping, or validation logic to sanitize data before it is re-injected into the agent context.
Audit Metadata