Get the login URL for the user. Use this to get the login URL for the user and then redirect the user to the login URL to get the request token.
AI agents call get_login_url to retrieve information from Zerodha without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves a login URL and performs no state-changing operations. It is purely informational: generating or fetching a URL for user authentication. While it is part of a financial trading platform (Zerodha), the tool itself only reads/returns data and does not move money, execute trades, or modify any account state. The subsequent user redirect is a client-side action outside the tool's scope.
From the tool's definition Tool description states 'Get the login URL for the user' — a retrieval operation that returns a URL string with no side effects or data modification.
Documented attack patterns abuse exactly the kind of access get_login_url gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Zerodha, and nothing reaches the server without passing your rules. This is the rule we recommend for get_login_url:
{
"version": "1",
"default": "deny",
"tools": {
"get_login_url": {}
}
}get_login_url 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.
Get the login URL for the user. Use this to get the login URL for the user and then redirect the user to the login URL to get the request token. It is categorised as a Read tool in the Zerodha MCP Server, which means it retrieves data without modifying state.
Register the Zerodha MCP server in PolicyLayer and add a rule for get_login_url: 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 Zerodha. Nothing to install.
get_login_url 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_login_url 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_login_url. 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_login_url is provided by the Zerodha MCP server (mtwn105/zerodha-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Zerodha, 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.
12 Zerodha tools catalogued and risk-classified — across an index of 43,000+ MCP servers.