memories
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval of stored memories which are then incorporated into the agent's active context. This creates a surface for indirect prompt injection if the stored content contains adversarial instructions intended to influence the agent's behavior.
- Ingestion points: Data retrieved via the retrieve-memory command is processed by the agent as shown in the examples in SKILL.md and scripts/memories.sh.
- Boundary markers: The skill instructions do not define delimiters or special instructions to isolate or ignore potentially malicious content within retrieved memories.
- Capability inventory: The skill has the capability to read from and write to a local embedding store via bun run commands in scripts/memories.sh.
- Sanitization: No sanitization or content validation logic is present in the bash wrapper or skill instructions.
- [COMMAND_EXECUTION]: The skill performs local command execution by using the bun runtime to execute workspace-specific tasks for memory management as defined in scripts/memories.sh.
Audit Metadata