docs_search
Search the Godot class reference (ClassDB) by keyword — matching class names and, unless a scope narrows it, their methods/properties/signals — and return each hit with its canonical docs URL. Read-only. Use kind to restrict to one member type, class_name to scope to a single class, and limit to ...
This record as markdown: /tools/breakpoint-mcp/docs-search.md
What docs_search does on Breakpoint
AI agents call docs_search to retrieve information from Breakpoint without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
deep | boolean | — | Also scan members, not just class names (default true) |
kind | string | — | Restrict to one result kind (default any) |
limit | integer | — | Max results before truncation (default 40) |
query | string | Yes | Case-insensitive substring to match against class / member names |
class_name | string | — | Scope the member search to a single class (still returns class-name matches project-wide) |
Parameters from the server's own tool schema.
Why docs_search is rated Low
This tool retrieves and queries documentation from the Godot class reference without modifying, executing, or deleting anything. The parameters (kind, class_name, limit) only narrow or filter the search results. There is no capability to modify state, execute code, or perform destructive actions.
From the tool's definition Tool explicitly describes itself as 'Read-only' and performs a search operation that 'return[s] each hit with its canonical docs URL' — a pure information retrieval action with no side effects.
Attacks that exploit this kind of access
The rule that runs docs_search safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Breakpoint, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For docs_search, this is the rule to start with:
docs_search is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Breakpoint, apply this rule, and every docs_search call is checked against it from then on.
Questions about docs_search
Search the Godot class reference (ClassDB) by keyword — matching class names and, unless a scope narrows it, their methods/properties/signals — and return each hit with its canonical docs URL. Read-only. Use kind to restrict to one member type, class_name to scope to a single class, and limit to bound results. It is categorised as a Read tool in the Breakpoint MCP Server, which means it retrieves data without modifying state.
docs_search accepts 5 parameters: deep, kind, limit, query, class_name. Required: query. The full parameter table on this page comes from the server's own tool schema.
Register the Breakpoint MCP server in PolicyLayer and add a rule for docs_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 Breakpoint. Nothing to install.
docs_search 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 docs_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.
Set action: deny in the PolicyLayer policy for docs_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.
docs_search is provided by the Breakpoint MCP server (breakpoint-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Breakpoint, and thousands of servers like it.
This server
Across the catalogue