Semantic (vector) search via RAG. Returns top-k chunks with paths and snippets. Optional
AI agents call similar_notes to retrieve information from brainMD without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a search/query tool that retrieves and returns data without side effects. It uses vector similarity to find related content but does not write, delete, execute code, or trigger external actions. The 'optional' parameter suggests flexible query constraints, but the fundamental operation remains read-only information retrieval.
From the tool's definition Tool performs 'semantic (vector) search via RAG' and 'returns top-k chunks with paths and snippets' - purely retrieval operation with no modification, deletion, or execution of code/external operations.
Documented attack patterns abuse exactly the kind of access similar_notes gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and brainMD, and nothing reaches the server without passing your rules. This is the rule we recommend for similar_notes:
{
"version": "1",
"default": "deny",
"tools": {
"similar_notes": {}
}
}similar_notes is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Semantic (vector) search via RAG. Returns top-k chunks with paths and snippets. Optional. It is categorised as a Read tool in the brainMD MCP Server, which means it retrieves data without modifying state.
Register the brainMD MCP server in PolicyLayer and add a rule for similar_notes: 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 brainMD. Nothing to install.
similar_notes is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the similar_notes 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 similar_notes. 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.
similar_notes is provided by the brainMD MCP server (mi4uu/brain.md). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from brainMD, 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.
17 brainMD tools catalogued and risk-classified — across an index of 43,000+ MCP servers.