linkedin-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@linkedin-mcpDraft a proposal to update my LinkedIn headline to 'Senior Product Manager'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
linkedin-mcp
MCP server for editing your own LinkedIn profile from Claude Code — with a hard propose-then-approve split: the model can draft a change and show you the diff, but only one tool can ever write, and it acts only on a proposal you have seen.
Partner gating: LinkedIn's Profile Edit API is restricted to LinkedIn-approved partner developers (
w_complianceis "a private permission and access is granted to select developers" — see docs/api-notes.md). Until partner approval lands,apply_proposalreturns an invalid-scope error. That is expected, not a defect. Everything else (OAuth, proposals, diffs) works today.
How it works
Tool | What it does | Writes to LinkedIn? |
| Prints the LinkedIn OAuth URL, catches the one-shot localhost redirect, exchanges the code, stores tokens | no (OAuth only) |
| Reports whether a token exists and when it expires | no |
| Fetches your profile ( | no |
| Builds the exact API request for a change (headline, summary, positions, skills, educations), saves it as a proposal, returns a unified diff + | never |
| Lists saved proposals | no |
| Deletes a saved proposal | no |
| Sends ONE saved proposal to LinkedIn — the only write tool. Code-enforced confirm gate: it refuses unless called with | yes |
Proposals persist under ~/.config/linkedin-mcp/proposals/ so an approval
can happen in a later session. Tokens and client credentials live in
~/.config/linkedin-mcp/.env with permissions 600 — entered by you, never
by an agent, never committed (see .env.example).
Install
Requires Python ≥ 3.11.
{
"mcpServers": {
"linkedin": {
"command": "bash",
"args": ["/path/to/linkedin-mcp/run.sh"]
}
}
}run.sh creates .venv/ and installs pinned dependencies on first launch
(stamp-gated; all bootstrap output goes to stderr, keeping the MCP stdio
channel clean).
Setup
Follow docs/SETUP-KYLE.md — it walks through creating
the LinkedIn Developer app, registering the redirect URL, filling
~/.config/linkedin-mcp/.env, running auth_start, and applying for the
partner program.
Development
bash run.sh --help # bootstraps the venv (runtime deps only)
.venv/bin/pip install -r requirements-dev.txt # adds pytest + coverage tooling
.venv/bin/python -m pytest # offline — every test runs against a mock transportThe test suite includes a granted-write fixture: a mock LinkedIn where
every documented write endpoint happily returns 200. Tests assert that
propose_edit leaves zero non-GET requests in the recorded log even
when writes would succeed, and (positive control) that apply_proposal does
record the documented write call in the same fixture. --live-probe is a
diagnostic flag (single real request, discriminates "endpoint right but
scope not granted" from "endpoint wrong"); it never runs in tests or CI and
requires LINKEDIN_MCP_LIVE_PROBE=1.
The API surface is pinned to dated verbatim excerpts from the official docs in docs/api-notes.md.
License
MIT — see LICENSE.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
233 tools for Google, Microsoft, TikTok, LinkedIn Ads in Claude or ChatGPT. Writes need approval.
Let AI tools securely access your LinkedIn network and DMs
Live SEO workflow tools for Claude Code, Codex, and AI agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/kyle-nelson-berkeley/linkedin-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server