LinkedIn is a professional networking platform enabling job seekers, companies, and thought leaders to connect, share content, and discover business opportunities
Verdict
Common use cases
- Draft and publish LinkedIn posts from briefs
- Delete outdated company announcements quickly
- Check posting permissions across managed pages
- Turn meeting notes into social updates
- Coordinate multi-account content calendars
Integration
- Vendor
- Category
- other
- Auth
- OAUTH2
- Tools
- 4
- Composio slug
linkedin
Tools
- Create a LinkedIn post
Creates a new post on linkedin for the authenticated user or an organization they manage; ensure the user has necessary permissions if posting for an organization.
- Delete LinkedIn Postdestructive
Deletes a specific linkedin post (share) by its unique `share id`, which must correspond to an existing share.
- Get company info
Retrieves organizations where the authenticated user has specific roles (acls), to determine their management or content posting capabilities for linkedin company pages.
- Get my info
Fetches the authenticated linkedin user's profile, notably including the 'author id' required for attributing content such as posts or articles.
Setup
Setup guide
- 11. Inside any Switchy Project, click the MCP icon in the toolbar and search for LinkedIn. 2. Click Connect and you'll be redirected to LinkedIn's OAuth consent screen. 3. Grant permissions for profile read access (r_liteprofile or equivalent) and content creation (w_member_social or w_organization_social if posting for company pages). 4. LinkedIn redirects you back to Switchy; the connection appears in your workspace's MCP list. 5. Open a Project and type '@LinkedIn get my info' to confirm the integration works — the AI should return your profile details including your author ID. 6. To post, use a prompt like '@LinkedIn create a post about our new product launch' and the AI will draft and publish it. 7. If you manage company pages, run '@LinkedIn get company info' first to see which organizations you can post for, then specify the organization in your post request.
What teammates see: by default, memories from LinkedIn are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Publish Product Update
@LinkedIn create a post announcing our new dashboard analytics feature. Highlight that it gives real-time insights and saves teams 5 hours per week. Keep it under 200 words and professional but approachable.Open in a Project →
Check Company Access
@LinkedIn get company info and show me which organizations I have posting permissions for. Format the results as a simple list with company names.Open in a Project →
Remove Old Post
@LinkedIn delete the post with share ID urn:li:share:7123456789. Confirm once it's removed.Open in a Project →
Fetch Author ID
@LinkedIn get my info and extract my author ID. Show just the ID in plain text.Open in a Project →
Weekly Thought Leadership
@LinkedIn create a post sharing insights from this article: [paste URL]. Focus on the main takeaway and add my perspective on why it matters for B2B SaaS teams. Aim for 150 words.Open in a Project →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@LinkedIn what organizations can I post content for, and what roles do I have at each?
Based on your LinkedIn permissions, you can post content for 3 organizations: 1. **Acme Corp** (ID: urn:li:organization:12345) — Administrator role, full posting rights 2. **Tech Startup Alliance** (ID: urn:li:organization:67890) — Content Manager, can create and schedule posts 3. **Industry Forum** (ID: urn:li:organization:54321) — Contributor, can draft posts for review You'll need the organization ID when creating posts on behalf of these pages. Your personal profile (ID: urn:li:person:abc123) can also publish directly.
This example uses the 'Get company info' tool to surface posting permissions. Useful before attempting to publish on behalf of a company page — the MCP returns organization IDs you'll reference in subsequent create-post calls. Requires OAuth scope for organization management.
@LinkedIn publish a post to my personal profile announcing our new product launch with a link to the landing page
Post created successfully on your LinkedIn profile. **Post ID:** urn:li:share:xyz789 **Author:** Your Name (urn:li:person:abc123) **Content:** "Excited to announce the launch of our new product! Check it out at https://example.com/product — we've been working on this for months and can't wait to hear your feedback." **Visibility:** Public **Posted:** Just now The post is now live on your profile and visible to your network.
This example demonstrates the 'Create a LinkedIn post' tool for personal profiles. The MCP first fetches your author ID via 'Get my info', then publishes. To post for an organization instead, you'd specify the org ID from the previous example. Rate limits apply — LinkedIn restricts post frequency per account.
@LinkedIn I posted something earlier today that has a typo — can you find my most recent post and delete it so I can repost?
Found your most recent post from 2 hours ago (ID: urn:li:share:xyz789) about the product launch. I've deleted it successfully. **Deleted post preview:** "Excited to announce the launch of our new product! Check it out at https://example.com/product..." You can now create a corrected version. Note that any engagement (likes, comments) on the original post is permanently lost — LinkedIn doesn't support editing posts after publication, only delete-and-repost.
This example chains 'Get my info' to retrieve recent activity, then 'Delete LinkedIn Post' to remove content. Highlights a LinkedIn platform limitation: no edit capability means typo fixes require deletion. The MCP needs the exact share ID, so it must first list or infer which post to target — deletion is irreversible.
Use-case deep-dives
When LinkedIn posting belongs in your AI workspace
A solo founder or exec assistant publishes one thought-leadership post per week, drafted in Switchy alongside customer notes and product updates. The LinkedIn MCP handles the final publish step without leaving the workspace. This works because the four-tool scope is narrow: get your author ID, post, delete if you typo, check org permissions if you manage a company page. The OAuth flow is one-time setup. If you're posting daily or running A/B tests on copy, you'll want a dedicated social tool with analytics. But for low-frequency, high-context posts where the draft lives in your AI workspace anyway, this MCP closes the loop without tab-switching.
Why this MCP doesn't scale to multi-client workflows
A three-person agency manages LinkedIn for eight B2B clients. Each client has its own company page, and posts go through a Notion approval queue before publish. The LinkedIn MCP can technically post to organizations the authenticated user manages, but OAuth ties you to one LinkedIn account at a time. Switching accounts mid-session breaks the flow, and there's no bulk-scheduling or draft-preview tooling in the four-tool set. If you're managing more than two company pages or need approval workflows with visual previews, a platform like Buffer or Hootsuite is the right call. This MCP is built for individual contributors publishing their own content, not agency-scale orchestration.
When LinkedIn posts aren't the integration you need
A two-person recruiting team sources candidates and tracks outreach in Airtable. They want to log LinkedIn activity alongside email threads and call notes. The LinkedIn MCP only touches posts and profile metadata—it doesn't read your feed, search for candidates, or pull message history. If your workflow is about content publishing (announcing a new role, sharing a company milestone), the MCP works. If you're trying to automate candidate research or track InMail threads, this isn't the integration. LinkedIn's API doesn't expose those surfaces to third-party MCPs. For recruiting workflows, you're better off with a dedicated ATS or a manual log-it-yourself system until LinkedIn opens more endpoints.
Frequently asked
What can the LinkedIn MCP do in Switchy?
It lets your team create and delete LinkedIn posts from Switchy's AI workspace, either on personal profiles or company pages you manage. You can also fetch profile details (like your author ID) and check which organizations you have posting rights for. It's built for content workflows where AI drafts posts and you publish them without leaving Switchy.
Do I need admin access to post on a company page?
You need specific LinkedIn roles that grant content posting permissions for that organization. The MCP includes a tool to check which companies you can post to before you try. If you're not an admin or content manager on the company page, the post will fail. Connect your personal LinkedIn account that has those roles.
Can it schedule posts or add images and videos?
No. The MCP creates text posts immediately when you run the tool—no scheduling, no media uploads. If you need rich media or delayed publishing, use LinkedIn's native scheduler or a dedicated social tool like Buffer. This integration is for fast text-only posting from AI-generated drafts.
How is this different from posting directly on LinkedIn?
It removes the copy-paste step. Your team can generate post ideas in Switchy, refine them with AI, and publish without opening LinkedIn. You lose LinkedIn's preview, hashtag suggestions, and media uploads, but you gain speed for text-first workflows. Best for teams that draft in AI and want one-click publishing.
Who on the team should connect this MCP?
Whoever owns your LinkedIn presence and has posting rights. For company pages, that's usually a marketing lead or social manager with the right LinkedIn role. Each Switchy user connects their own LinkedIn account via OAuth, so posts appear under their name or their managed organizations—not a shared bot account.