AI agents use backup_restore to create or update resources in Redis MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Redis MCP environment.
Backup restoration is a write operation that overwrites the current database contents with previous state. While technically reversible (the current state could be backed up first), it causes significant data modification and potential data loss if the wrong backup is restored. Not Destructive because restoration doesn't permanently delete data from all backups; not Execute because it doesn't run arbitrary code.
From the tool's definition Tool name 'backup_restore' and description '从备份恢复 Redis 数据' (restore Redis data from backup) indicate this performs data restoration—a write operation that modifies the current database state by overwriting it with backed-up data.
Attacks that exploit this kind of access
从备份恢复 Redis 数据. It is categorised as a Write tool in the Redis MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Redis MCP server in PolicyLayer and add a rule for backup_restore: 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 Redis MCP. Nothing to install.
backup_restore 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 backup_restore 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 backup_restore. 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.
backup_restore is provided by the Redis MCP server (pickstar-2002/redis-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →