Skip to content

ROLEBUG MCP SERVER

Let your AI agent move your job search forward

Find marketing roles, review your fit and prepare tailored CVs and cover letters through the agent you already use.

Server URL
https://rolebug.com/mcp

Streamable HTTP · OAuth sign-in · Included with Pro

ClaudeChatGPTPerplexityHermesOpenClawClaude CodeCodexCursorGemini CLI
Connect your agent

SET UP ONCE, PICK UP WHERE YOU LEFT OFF

Connect RoleBug to your agent

Start with a RoleBug account, complete onboarding and activate Pro. Then add the server in your agent and sign in when it asks.

Set up your RoleBug account ↗

Connect Claude and Cowork

Custom connector

  1. Add a custom connector

    In Claude, open Customize → Connectors → Add custom connector. Name it RoleBug and paste the server URL.

    https://rolebug.com/mcp

  2. Connect your account

    Select Connect, sign in to RoleBug and approve the permissions you want to allow. Leave optional OAuth client ID and secret fields empty.

  3. Enable RoleBug

    In a new Claude conversation, enable RoleBug from the connectors menu. Use the same remote connector in Cowork.

Need help connecting?

Missing the connector? Check that you added it in the workspace you are using. Team workspaces may require an administrator to enable custom connectors.

Client documentation ↗

Connect ChatGPT

Custom MCP connector

  1. Open app settings

    Open Plugins in ChatGPT, select the plus button and choose Add custom MCP server. Account and workspace policies may limit this option.

  2. Create the connection

    Name it RoleBug. Under Connection, choose a public endpoint and enter this URL. Configure OAuth authentication, review the warning and select Create as a plugin.

    https://rolebug.com/mcp

  3. Sign in and enable RoleBug

    Complete RoleBug sign-in and approve access. Install the resulting plugin, start a new chat, type @ and select RoleBug.

Need help connecting?

RoleBug is added manually for now. There is no published RoleBug directory listing. If Add custom MCP server is unavailable, check your account and workspace permissions.

Client documentation ↗

Connect Perplexity

Remote connector

  1. Open Connectors

    Open Plugins → Add custom → Connector → Add MCP connector, then choose Remote. In older layouts, use Customize → Connectors or Account settings → Connectors.

  2. Add RoleBug

    Name it RoleBug, choose OAuth authentication and Streamable HTTP transport, then paste this URL.

    https://rolebug.com/mcp

  3. Authorize the connector

    Save it, open the RoleBug connector card and finish RoleBug sign-in. Enable it for your conversation or Computer workflow.

Need help connecting?

If Custom connector is missing, check your Perplexity plan or ask your workspace administrator to allow remote connectors.

Client documentation ↗

Connect Hermes

Configuration + terminal sign-in

Let your agent set it up

Copy this into Hermes. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in Hermes using its remote MCP server.

1. Read ~/.hermes/config.yaml and preserve the existing configuration. If rolebug is configured with a different URL, ask before replacing it.
2. Merge this entry under mcp_servers without duplicating that key:
   rolebug:
     url: https://rolebug.com/mcp
     auth: oauth
3. Run hermes mcp login rolebug in a separate terminal process. Keep it running and send me the authorization URL, or tell me to finish in the browser that opens. Wait for me to approve access. If Hermes runs on a remote host, explain its paste-back or SSH callback step; do not claim approval alone completed login.
4. Tell me to use /reload-mcp or start a fresh Hermes session if needed to load the tools.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If you cannot edit the Hermes configuration or run its CLI here, stop and tell me to use the manual setup instructions.

  1. Add RoleBug to Hermes

    Merge this entry into ~/.hermes/config.yaml. Keep any existing mcp_servers entries.

    mcp_servers:
  rolebug:
    url: https://rolebug.com/mcp
    auth: oauth

  2. Complete browser sign-in

    Run this in a fresh terminal, then follow the authorization link and approve RoleBug access.

    hermes mcp login rolebug

  3. Reload your session

    Start a fresh Hermes session or reload its MCP tools.

    /reload-mcp
Need help connecting?

On a remote host, use Hermes’s paste-back or SSH callback instructions. RoleBug does not currently offer device-code login.

Client documentation ↗

Connect OpenClaw

Terminal + browser sign-in

Let your agent set it up

Copy this into OpenClaw. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in OpenClaw using its remote MCP server.

1. Check that this OpenClaw version supports native MCP OAuth. Preserve my other MCP servers. If rolebug already points somewhere else, ask before replacing it.
2. If missing, run:
openclaw mcp set rolebug '{"url":"https://rolebug.com/mcp","transport":"streamable-http","auth":"oauth"}'
3. Run openclaw mcp login rolebug on the machine running OpenClaw. Keep it running and send me the authorization URL. Wait for me to approve access. For a remote host, follow OpenClaw's manual callback instructions if needed.
4. Run openclaw mcp reload, then openclaw mcp probe rolebug --json. Confirm the active agent runtime can access the tools; a saved configuration alone is not a working connection.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If native MCP OAuth or CLI access is unavailable, stop and tell me to use the manual setup instructions.

  1. Save the remote server

    Use an up-to-date OpenClaw installation with native MCP OAuth support. Run this on the machine that runs OpenClaw.

    openclaw mcp set rolebug '{"url":"https://rolebug.com/mcp","transport":"streamable-http","auth":"oauth"}'

  2. Authorize RoleBug

    Run the command and open the authorization link. If OpenClaw is on a server, follow its manual callback instructions.

    openclaw mcp login rolebug

  3. Reload and verify

    Reload the server configuration, then probe the connection before asking your agent to use it.

    openclaw mcp reload
openclaw mcp probe rolebug --json
Need help connecting?

OpenClaw runtimes differ in how they expose saved MCP servers. A successful probe confirms access, but check that your active agent runtime actually has the RoleBug tools. Older builds may need updating.

Client documentation ↗

Connect Claude Code

Terminal + browser sign-in

Let your agent set it up

Copy this into Claude Code. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in Claude Code using its remote MCP server.

1. Check whether rolebug is already configured. Preserve my other MCP servers; if this name points somewhere else, ask before replacing it.
2. If missing, run: claude mcp add --transport http --scope user rolebug https://rolebug.com/mcp
3. Run: claude mcp login rolebug. Keep the login process running. Send me the authorization URL if it is printed, or tell me to finish in the browser that opens. Wait for me to approve access. If this Claude Code version has no mcp login command, tell me to open /mcp, select rolebug and authenticate there.
4. If a new conversation is needed to load the tools, explain that handoff and give me the check prompt below.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If you cannot run the Claude Code CLI here, stop and tell me to use the manual setup instructions.

  1. Add RoleBug in your terminal

    Run this command where Claude Code is installed. User scope makes it available across your projects.

    claude mcp add --transport http --scope user rolebug https://rolebug.com/mcp

  2. Sign in to RoleBug

    Run this command and finish sign-in in your browser. On older Claude Code versions without this command, open /mcp inside Claude Code and authenticate rolebug there.

    claude mcp login rolebug

  3. Start a fresh conversation

    Once rolebug shows as connected, start a new conversation and try the connection check below.

Need help connecting?

Tools missing or access expired? Open /mcp, authenticate rolebug again and start a fresh conversation.

Client documentation ↗

Connect Codex

Terminal + browser sign-in

Let your agent set it up

Copy this into Codex. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in Codex using its remote MCP server.

1. Check whether rolebug is already configured. Preserve my other MCP servers; if this name points somewhere else, ask before replacing it.
2. If missing, run: codex mcp add rolebug --url https://rolebug.com/mcp
3. Run: codex mcp login rolebug. Keep the login process running and send me the authorization URL if it is printed, or tell me to finish in the browser that opens. Wait for me to approve access.
4. Check the connection. If the current session cannot load the new tools, tell me to open a new Codex session and give me the check prompt below to continue there.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If you cannot run the Codex CLI here, stop and tell me to use the manual setup instructions.

  1. Add the MCP server

    Run this in your terminal. It adds RoleBug to the MCP configuration used by your local Codex clients.

    codex mcp add rolebug --url https://rolebug.com/mcp

  2. Sign in to RoleBug

    Run the login command, then complete sign-in and approve permissions in your browser.

    codex mcp login rolebug

  3. Open a new Codex session

    Restart the session so Codex loads the RoleBug tools. You can check the server in /mcp.

Need help connecting?

You can also add a Streamable HTTP server in the desktop app’s MCP settings and select Authenticate. If access expires, run codex mcp login rolebug again.

Client documentation ↗

Connect Cursor

MCP settings / configuration

Let your agent set it up

Copy this into Cursor. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in Cursor using its remote MCP server.

1. Read ~/.cursor/mcp.json. Preserve existing entries and ask before replacing a rolebug entry that points elsewhere.
2. Merge this entry under mcpServers without duplicating that key:
   "rolebug": {"url": "https://rolebug.com/mcp"}
3. Tell me to open Cursor's MCP settings, enable rolebug and select its Connect or sign-in action. Wait for me to finish RoleBug sign-in in my browser and approve access.
4. Ask me to open a new Agent conversation if needed to load the tools, and give me the check prompt below.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If you cannot edit Cursor's MCP configuration here, stop and tell me to use the manual setup instructions.

  1. Add the server configuration

    Open Cursor’s MCP settings, or merge this entry into ~/.cursor/mcp.json for all projects. Keep any existing server entries.

    {
  "mcpServers": {
    "rolebug": {
      "url": "https://rolebug.com/mcp"
    }
  }
}

  2. Connect your account

    In Cursor’s MCP settings, enable rolebug and select Connect or the sign-in action. Finish RoleBug sign-in and approve permissions.

  3. Use it in Agent chat

    Start a new Agent conversation and ask it to check RoleBug. Approve tool requests when Cursor asks.

Need help connecting?

Check rolebug is enabled in MCP settings. If authentication is requested again, reconnect there and start a new Agent conversation.

Client documentation ↗

Connect Grok Bot

Custom MCP connector · Setup in testing

Before you connect

Make sure Grok Bot has active access on your account before connecting. Eligible plans include Cursor Premium or higher, legacy paid Cursor plans, or a linked X Premium+ or SuperGrok subscription. These are separate from RoleBug Pro.

View supported Grok Bot plans ↗

  1. Open Plugins

    In the Grok Bot app, look under Settings → Plugins for the option to add a custom MCP server. This is separate from Grok chat.

  2. Add RoleBug

    Name the server RoleBug and enter this URL. Use browser sign-in if the app offers it.

    https://rolebug.com/mcp

  3. Finish sign-in

    Sign in to RoleBug, approve access, then attach RoleBug to a task from the @ menu. This client flow still needs testing.

Need help connecting?

These steps follow Adkit’s Grok Bot guide. RoleBug compatibility has not been verified in the app; the setup may change after testing.

Reference guide ↗

Connect Meta Muse

Connect through chat · Setup in testing

  1. Ask Muse to add RoleBug

    Paste this message into a new Muse conversation.

    Add RoleBug as a custom MCP connector at https://rolebug.com/mcp. It uses Streamable HTTP and OAuth browser sign-in, with no API key. Send me the authorization link and wait for me to approve access.

  2. Approve access

    Open the authorization link, sign in to RoleBug and choose the permissions to allow.

  3. Check the connection

    Ask Muse to run the read-only connection check below. This custom connector flow still needs testing.

Need help connecting?

These steps follow Adkit’s Muse guide. RoleBug has no published Muse directory listing, and the chat setup has not been verified in Muse.

Reference guide ↗

Connect Gemini CLI

Terminal + browser sign-in

Let your agent set it up

Copy this into Gemini CLI. You finish RoleBug sign-in in your browser.

Help me set up RoleBug in Gemini CLI using its remote MCP server.

1. Check whether rolebug is already configured. Preserve my other MCP servers; ask before replacing a different URL.
2. If missing, run: gemini mcp add --transport http rolebug https://rolebug.com/mcp
3. Tell me to enter /mcp auth rolebug inside Gemini CLI. This is an interactive Gemini command, not a shell command. Wait for me to finish browser sign-in and approve access.
4. Tell me to reload the tools or start a fresh session if needed. Do not report success just because the server was added.
5. Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything. If my profile is empty, say so.
6. Only report success after a RoleBug tool call succeeds. Otherwise explain what failed and the next step.

If you cannot run Gemini CLI here, stop and tell me to use the manual setup instructions.

  1. Add RoleBug

    Run this in your terminal with Gemini CLI installed.

    gemini mcp add --transport http rolebug https://rolebug.com/mcp

  2. Authorize inside Gemini CLI

    Start Gemini CLI and run this slash command. Complete RoleBug sign-in in your browser.

    /mcp auth rolebug

  3. Check the available tools

    Use /mcp to check the connection, then try the prompt below in a new conversation.

Need help connecting?

Adding the URL does not finish sign-in. Run /mcp auth rolebug if tools are missing or you see an authentication error.

Client documentation ↗

Connect n8n

Workflow node + OAuth · Setup in testing

  1. Add an MCP Client Tool

    Connect an MCP Client Tool node to your AI Agent node. Choose HTTP Streamable and enter this endpoint.

    https://rolebug.com/mcp

  2. Use an OAuth connection

    RoleBug uses browser sign-in, not an API key. The OAuth credential setup and refresh behavior for n8n still need validation before this guide is ready.

  3. Wait for the verified setup

    We will add the exact credential fields and connection steps once this flow has passed testing.

Need help connecting?

If n8n asks for a bearer token, do not paste a RoleBug session cookie or copy a temporary token. A verified OAuth credential guide is still needed.

Reference guide ↗

Check that you’re connected

Try this in your agent after signing in.

Use RoleBug to check my connection and available permissions, then show my profile overview. Do not change anything.

YOUR FIRST CONVERSATION

Start with something useful

Copy a prompt into your connected agent. Replace anything in square brackets with your details.

Know where you stand

Use RoleBug to show my experience, target roles and anything missing from my profile. Do not change anything.

Your saved experience, ready to work from

Find roles worth a closer look

Find remote growth marketing roles in RoleBug. Show me the location requirements and explain which ones look relevant to my saved experience.

A shortlist with the details that matter

See how you fit

Review this role against my saved experience: [job URL]. Explain my strengths, the gaps and what I would need to demonstrate.

A role-fit review before you spend time applying

Prepare a tailored application

Prepare a CV and cover letter for [saved role] using my saved experience. Show me the drafts before saving.

Documents built around the role and your experience

Get your CV ready to share

Export my saved CV for [saved role] as PDF and Word.

Your saved CV in a RoleBug template

Keep your next step in sight

I have an interview for [saved role] next Tuesday. Update my application history and remind me to prepare on Monday.

Application updates and a reminder in your workspace

THE SAME WORKSPACE, ANOTHER WAY TO WORK

What your agent can do

Use your experience
Work from the career history, strengths, preferences and materials saved in your RoleBug profile.
Find and review marketing roles
Search the RoleBug job directory, save jobs you find elsewhere and compare a role’s requirements with your experience.
Prepare and revise documents
Create tailored CVs and cover letters, revise the wording and export a saved CV as PDF or Word in a RoleBug template.
Keep track of your progress
Record applications, replies and interviews, add follow-up reminders and continue from the same history in RoleBug.

Send applications and messages yourself. Use Max Mode in the RoleBug workspace.

YOUR ACCOUNT, YOUR CALL

You choose what your agent can access

01

Sign in to your account

RoleBug opens in your browser. Use the same account as your workspace. There’s no API key to copy.

02

Choose permissions

Review what the agent wants to read or change before approving the connection.

03

Stay in control

Your work stays in RoleBug. Revoke access anytime from Settings → Connected agents.

A FEW USEFUL DETAILS

Frequently asked questions

What is the RoleBug MCP server?

It connects your AI agent to your RoleBug account using the Model Context Protocol. Your agent can use your saved experience, search marketing jobs, review roles, prepare documents and update your application history.

Do I need RoleBug Pro?

Yes. Create a RoleBug account, finish onboarding and activate Pro before connecting. MCP access is included in Pro. Your agent provider may have its own plan requirements.

Do I need an API key?

No. Add the server URL, then sign in to RoleBug in your browser and approve access.

Will I see the same work in RoleBug?

Yes. Your agent uses the experience, roles, documents and progress saved in your account. You can continue working in the RoleBug workspace.

Can my agent submit applications for me?

RoleBug can prepare documents and record the application updates you provide. It does not send messages or submit applications to employers.

Can I use Max Mode through my agent?

Max Mode is currently available in the RoleBug workspace. The MCP tools cover job search, role reviews, application documents and tracking.

How do I disconnect an agent?

Open Settings → Connected agents in RoleBug and revoke the connection. Reconnect from your agent if you want to allow access again.

Keep your next move moving

Connect once. Work from your saved experience, roles and progress in the agent you already use.

Open agent setup