MCP for Claude Code, Codex, and ChatGPT

The LinkedIn MCP server for warm leads

Signado connects your existing workspace to Claude Code, Codex, and ChatGPT through MCP, so you can list warm leads, show the LinkedIn activity that made them relevant, find emails, and save AI outreach drafts back to Campaigns. Sign in once with OAuth, no API keys to manage.

production E2E facts

> list my newest warm leads and show the LinkedIn activity I can reference

list_warm_leads returned 5 real warm leads with icp_score and engagements.

preview_source charged exactly 1 credit and returned posts with engagers.

Need a draft next? draft_message costs 7 credits per contact and saves to Campaigns.

Claude Code, Codex, and ChatGPT all connect over OAuth and call your tools. No API keys, same credit pool as the app.

Connect it to your AI tools

Endpoint: https://mcp.signado.io/mcp. Use your existing account, then choose the workspace to connect.

Claude Code

claude mcp add --transport http signado https://mcp.signado.io/mcp

Then run /mcp, choose signado, and authenticate in your browser.

Codex

codex mcp add signado --url https://mcp.signado.io/mcp
codex mcp login signado

The login opens a browser approval flow. Approve it, then ask Codex to call ping.

ChatGPT

https://mcp.signado.io/mcp

Settings, Apps, Advanced settings, turn on developer mode. Apps, Create, paste the endpoint above, choose OAuth. Approve the sign-in and pick your workspace.

OpenClaw

https://mcp.signado.io/mcp

Add Signado as a streamable-http MCP server, run openclaw mcp login, and approve the sign-in in your browser.

Works with any MCP client

Signado speaks standard MCP, so Cursor, VS Code, Gemini CLI, and anything else that supports it connect the same way: point them at the endpoint and sign in. Same OAuth, same credit pool, no API keys to copy or rotate.

Need screenshots or fallbacks? Use the setup guide.

Every tool, same credit pool

No connection fee. Free tools read or organize your workspace. Paid tools return credits charged, balance after, and session spend total in the response.

ToolWhat it doesCost
Read and export
pingCheck that MCP is connected to your workspace.0 credits
list_warm_leadsList warm leads with ICP scores, companies, and LinkedIn engagement context.0 credits
get_warm_lead_detailLoad one warm lead’s complete LinkedIn engagement history on demand.0 credits
export_leadsExport warm leads as structured CSV rows for a date window.0 credits
list_templatesList system templates, workspace templates, and authoring rules.0 credits
Discovery sources
preview_sourcePreview posts and engagers for a keyword or creator before activation.1 credit
add_sourceAdd a monitoring source: track a competitor or creator LinkedIn profile, or a LinkedIn keyword. Shows projected daily credit cost before activation and requires confirmation.0 credits to add, 3 credits per new warm lead
run_source_onceRun an active or paused keyword, creator, or LinkedIn post source once, with a credit ceiling you confirm first.0 credits to start, 3 credits per new warm lead found
cancel_source_runCancel an in-flight one-time source run while keeping leads already found.0 credits
catchup_sourceLegacy alias of run_source_once, kept for backward compatibility.0 credits to start, 3 credits per new warm lead found
remove_sourceArchive a source while keeping the history in place.0 credits
Enrichment and outreach
find_emailFind emails for selected leads and write results back to Warm Leads.5 credits per contact
qualifyPull each selected lead's company data into a quick fit check before outreach.2 credits per contact
draft_messageCreate campaign-backed AI drafts for selected warm leads.7 credits per contact
save_templateSave a workspace template with custom variables for future drafts.0 credits

OAuth, no API keys

MCP connects through browser sign-in. Access is scoped to the workspace you choose.

No keys to manage

There is nothing to copy into config, rotate, or keep out of a repository. Sign-in happens in your browser.

Workspace-scoped

MCP reaches the workspace you approved. The app's row-level security applies to every call.

Receipts on paid calls

Every paid response includes credits charged, balance after, and session spend total.

FAQ

Practical answers

Does MCP access cost extra?

No. Every client uses the same credit pool and the same rates as the app. Reading leads is free, source previews cost 1 credit, email lookups cost 5 credits per contact, Qualify costs 2 credits per contact, AI drafts cost 7 credits per contact, and discovery charges 3 credits per new warm lead.

Which AI tools does it work with?

Claude Code, Codex, and ChatGPT are tested and ready. ChatGPT connects through developer mode. Because Signado speaks standard MCP, Cursor, OpenClaw, VS Code, Gemini CLI, and any other MCP client connect the same way.

Is this a LinkedIn MCP server?

Yes, in the sense that matters: it gives your AI tools the warm leads Signado finds from LinkedIn engagement, scored for fit, with what each person said. It never logs into your LinkedIn account, so there is nothing to get your account restricted.

Do I need an API key?

No. Sign in with your existing account, choose the workspace to connect, and approve the OAuth prompt in your browser.

What can MCP access?

Only the workspace you selected during sign-in. The same app permissions and row-level security apply to every tool call.

Where do results land?

Warm leads stay on Warm Leads, found emails write back to the lead rows, source changes appear in Settings, and AI drafts are saved to Campaigns.

What happens if there is an issue?

The MCP help page has the current setup checklist, known client quirks, and troubleshooting steps. Open the MCP help page.

Connect it, then ask for your newest warm leads

The Free plan includes 200 lifetime credits. Use them from the app, Claude Code, or Codex.