manual_login

THE RISK

Medium Risk

Why manual_login needs a policy

This tool modifies persistent authentication state by saving session cookies, enabling future tool executions without re-authentication. While not destructive or financial in isolation, it is Write-category because it creates/modifies stored data (session state).

From the tool's definition Saves Upwork session cookies (persistent state modification) via Chrome DevTools Protocol connection to an existing browser tab. Described as 'save Upwork session cookies' which creates and stores authentication state.

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

POLICY

How to control manual_login

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

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "manual_login": {
      "limits": [
        {
          "counter": "manual_login_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

manual_login stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

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

LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Upwork tools

Write save_session Alias for manual_login — re-save session from current Chrome tab. Write send_message Send a message in an Upwork conversation. Write submit_proposal Submit a proposal for an Upwork job. Call get_job_details first to get screening questions Write update_profile Update your Upwork profile fields: title, description/bio, hourly_rate, skills. Read analyze_job Score a job opportunity 0-100 across 5 dimensions: niche fit, client quality, budget fit, Read get_job_details Get full job details: description, screening questions, client history, budget. Call this Read get_messages Read Upwork messages. Without room_id: lists conversations. With room_id: reads full messa Read get_profile Get your Upwork freelancer profile, title, skills, hourly rate, JSS score, connects balanc

All 11 Upwork tools →

Write tools on other servers

M-Team MCP Server download_torrent YouTube MCP Server generate_video_title Android Forensics ADB MCP Server combine_reports 0xarchive web3_signup

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Argument validation → Checking tool call arguments against schema and policy before they run.
Scoped token → Per-person credentials that grant a defined subset of servers and tools.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about manual_login

What does the manual_login tool do? +

Connect to existing Chrome tab via CDP and save Upwork session cookies. Run connect-chrome.bat first, then call this. It is categorised as a Write tool in the Upwork MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on manual_login? +

Register the Upwork MCP server in PolicyLayer and add a rule for manual_login: 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 Upwork. Nothing to install.

What risk level is manual_login? +

manual_login is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit manual_login? +

Yes. Add a rate_limit block to the manual_login 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 manual_login completely? +

Set action: deny in the PolicyLayer policy for manual_login. 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 manual_login? +

manual_login is provided by the Upwork MCP server (zcrossoverz/upwork-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Upwork tool call.

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

CHECK YOUR STACK →