kg_search

THE RISK

Low Risk

Why kg_search needs a policy

This tool retrieves or queries data from a knowledge graph without producing side effects. It supports searching across knowledge nodes but does not create, modify, delete, or execute operations. The server's purpose is to capture and preserve decision context, and this tool serves the read/retrieval function within that system. No destructive, financial, or code execution capabilities are indicated.

From the tool's definition Tool name 'kg_search' and description explicitly states it is a 'search tool for finding knowledge nodes' with no mention of modification, deletion, execution, or financial operations. The verb 'finding' and 'search' indicate retrieval/query operations only.

Documented attack patterns abuse exactly the kind of access kg_search gives an agent:

POLICY

How to control kg_search

PolicyLayer is an MCP gateway — it sits between your AI agents and Personal Kg, and nothing reaches the server without passing your rules. This is the rule we recommend for kg_search:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "kg_search": {}
  }
}

kg_search is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.

Create a free account and register Personal Kg — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

CAP THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Personal Kg tools

Execute kg_diagnostic Generate a diagnostic HTML dashboard for debugging knowledge graph context selection, sear Write kg_capture Primary tool for capturing knowledge nodes. Use this to record decisions, progress updates Write kg_capture_session Captures session summaries with structured metadata. Use at the end of work sessions to re Write kg_link_session Creates a relationship between a session node and another knowledge node. Use to explicitl Write kg_resolve_question Marks a question as resolved by linking it to a decision, insight, or other node that answ Write kg_update_node Updates an existing knowledge node. Use to modify content, tags, importance, or visibility Read kg_session_warmup Start every session with this tool! Loads comprehensive context about your project includi Other kg_admin Unified admin tool for maintenance operations. Supports:

All 18 Personal Kg tools →

Read tools on other servers

M-Team MCP Server get_torrent_detail YouTube MCP Server search_channel Android Forensics ADB MCP Server parse_browser_history 0xarchive web3_challenge

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Data exfiltration → How read access becomes an exfiltration channel when chained with untrusted content.
MCP token cost → What connecting this server costs in context-window tokens on every request.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about kg_search

What does the kg_search tool do? +

Unified search tool for finding knowledge nodes. Supports three modes:. It is categorised as a Read tool in the Personal Kg MCP Server, which means it retrieves data without modifying state.

How do I enforce a policy on kg_search? +

Register the Personal Kg MCP server in PolicyLayer and add a rule for kg_search: 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 Personal Kg. Nothing to install.

What risk level is kg_search? +

kg_search is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit kg_search? +

Yes. Add a rate_limit block to the kg_search 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.

How do I block kg_search completely? +

Set action: deny in the PolicyLayer policy for kg_search. 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.

What MCP server provides kg_search? +

kg_search is provided by the Personal Kg MCP server (tomschell/personal-kg-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Personal Kg tool call.

Start from Personal Kg, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →