semantic_search

THE RISK

Low Risk

Why semantic_search needs a policy

The tool retrieves and queries code data through semantic search without side effects, fitting the Read category. Severity is elevated to high because the underlying codebase knowledge graph may contain sensitive architectural information, trade secrets, or confidential implementation details.

From the tool's definition Tool performs 'Search for code using natural language' with no indication of modification, deletion, or execution. Described as a search operation, which is read-only.

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

POLICY

How to control semantic_search

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

policy.json

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

semantic_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 GraphHub — 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 GraphHub tools

Destructive forget Delete observations from memory. Can delete by ID, session, project, type, or date. Execute query_graph Run a direct Cypher query against the codebase graph database. Use this to find complex re Write remember_bugfix Record a resolved bug so future agents can reuse the fix pattern. Stores symptom, root cau Write remember_skill_choice Cache which .skill.md / SKILL.md file was used for a task so similar future tasks route to Write remember Save a learning, decision, finding, or other observation to persistent memory. Searchable Write update_observation Update an existing observation Read batch_context Get definition + caller/callee counts for multiple symbols in one call. Replaces N get_con Read changed_symbols List the symbols in files changed by the current branch (or working tree / staged diff). E

All 32 GraphHub 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 semantic_search

What does the semantic_search tool do? +

Search for code using natural language. Supports semantic (meaning-based), keyword, or hybrid search modes. It is categorised as a Read tool in the GraphHub MCP Server, which means it retrieves data without modifying state.

How do I enforce a policy on semantic_search? +

Register the GraphHub MCP server in PolicyLayer and add a rule for semantic_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 GraphHub. Nothing to install.

What risk level is semantic_search? +

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

Can I rate-limit semantic_search? +

Yes. Add a rate_limit block to the semantic_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 semantic_search completely? +

Set action: deny in the PolicyLayer policy for semantic_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 semantic_search? +

semantic_search is provided by the GraphHub MCP server (slnquangtran/graph-hub). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every GraphHub tool call.

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

CHECK YOUR STACK →