AI agents use restore_backup to create or update resources in Claude Code Toolkit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Code Toolkit environment.
Restoring a backup is a reversible write operation that modifies data (conversation files) but does not permanently delete data. It has high severity because restoring the wrong backup could overwrite current conversation state, potentially losing recent work or context, but remains Write rather than Destructive because the operation is theoretically reversible (the current state could be backed up before restore).
From the tool's definition Tool name 'restore_backup' and description 'Restore a conversation file from a backup' indicate the tool modifies existing data by restoring/overwriting conversation files with backup versions.
Documented attack patterns abuse exactly the kind of access restore_backup gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Code Toolkit, and nothing reaches the server without passing your rules. This is the rule we recommend for restore_backup:
{
"version": "1",
"default": "deny",
"tools": {
"restore_backup": {
"limits": [
{
"counter": "restore_backup_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}restore_backup stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Restore a conversation file from a backup. It is categorised as a Write tool in the Claude Code Toolkit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Code Toolkit MCP server in PolicyLayer and add a rule for restore_backup: 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 Claude Code Toolkit. Nothing to install.
restore_backup 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 restore_backup 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 restore_backup. 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.
restore_backup is provided by the Claude Code Toolkit MCP server (asifkibria/claude-code-toolkit). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claude Code Toolkit, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
40 Claude Code Toolkit tools catalogued and risk-classified — across an index of 43,000+ MCP servers.