Re-fetch the OpenAPI spec from LNbits and rebuild the tool list.
AI agents invoke refresh_tools to trigger actions in LNbits MCP Server. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool triggers an external operation (fetching from LNbits) and rebuilds/modifies the active tool list, which is a dynamic execution action with side effects on the agent's available capabilities. It is not purely a read since it 'rebuilds' the tool list, but it is not financial or destructive. Execute is the most appropriate category given it triggers an external fetch and reconstructs internal state.
From the tool's definition Re-fetch the OpenAPI spec from LNbits and rebuild the tool list
Documented attack patterns abuse exactly the kind of access refresh_tools gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and LNbits MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for refresh_tools:
{
"version": "1",
"default": "deny",
"tools": {
"refresh_tools": {
"limits": [
{
"counter": "refresh_tools_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}refresh_tools stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Re-fetch the OpenAPI spec from LNbits and rebuild the tool list. It is categorised as a Execute tool in the LNbits MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the LNbits MCP Server MCP server in PolicyLayer and add a rule for refresh_tools: 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 LNbits MCP Server. Nothing to install.
refresh_tools is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the refresh_tools 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 refresh_tools. 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.
refresh_tools is provided by the LNbits MCP Server MCP server (lnbits/lnbits-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from LNbits MCP Server, 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.
6 LNbits MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.