Sends a complex reasoning task to the best available local/network LLM.
AI agents invoke query_local_model to trigger actions in M3 Memory. 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 execution on an external or local LLM with an arbitrary reasoning task. The effects depend entirely on the content of the task passed as arguments, and it involves triggering external computation/operations. This falls under Execute since it runs operations on external systems (LLM inference endpoints).
From the tool's definition Sends a complex reasoning task to the best available local/network LLM
Documented attack patterns abuse exactly the kind of access query_local_model gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and M3 Memory, and nothing reaches the server without passing your rules. This is the rule we recommend for query_local_model:
{
"version": "1",
"default": "deny",
"tools": {
"query_local_model": {
"limits": [
{
"counter": "query_local_model_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}query_local_model 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.
Sends a complex reasoning task to the best available local/network LLM. It is categorised as a Execute tool in the M3 Memory MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the M3 Memory MCP server in PolicyLayer and add a rule for query_local_model: 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 M3 Memory. Nothing to install.
query_local_model 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 query_local_model 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 query_local_model. 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.
query_local_model is provided by the M3 Memory MCP server (skynetcmd/m3-memory). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from M3 Memory, 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.
43 M3 Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.