AI agents call get_tasks to retrieve information from Twenty MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves task data from the CRM system. The verb 'Retrieve' and the absence of any language suggesting modification, deletion, or execution of operations confirms this is a read-only operation. The blast radius is minimal as it only accesses data the authenticated user should already be able to view through normal CRM access.
From the tool's definition Tool name 'get_tasks' and description 'Retrieve tasks from Twenty CRM' indicate a query operation that retrieves existing data without modification or side effects.
Documented attack patterns abuse exactly the kind of access get_tasks gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Twenty MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for get_tasks:
{
"version": "1",
"default": "deny",
"tools": {
"get_tasks": {}
}
}get_tasks 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.
Retrieve tasks from Twenty CRM. It is categorised as a Read tool in the Twenty MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Twenty MCP Server MCP server in PolicyLayer and add a rule for get_tasks: 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 Twenty MCP Server. Nothing to install.
get_tasks 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 get_tasks 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 get_tasks. 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.
get_tasks is provided by the Twenty MCP Server MCP server (jezweb/twenty-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Twenty MCP Server, 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.
29 Twenty MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.