AI agents use save_memory to create or update resources in Hi-AI — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Hi-AI environment.
An AI agent can call save_memory faster than any human can review — one bad instruction and it creates or modifies resources in Hi-AI by the hundred, each call as confident as the last.
Attacks that exploit this kind of access
중요한 정보를 장기 메모리에 저장합니다. 프로젝트 결정사항, 아키텍처, 설정 등을 기록하세요. 키워드: 기억해, remember, 저장해, save, memorize, keep 💡 저장 후 link_memories로 관련 메모리를 연결하면 지식 그래프가 구축됩니다. It is categorised as a Write tool in the Hi-AI MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Hi-AI MCP server in PolicyLayer and add a rule for save_memory: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Hi-AI. Nothing to install.
save_memory is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the save_memory rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for save_memory. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
save_memory is provided by the Hi-AI MCP server (su-record/hi-ai). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.