LinkedIn Community Management MCP connector for Copilot Studio

March 10, 2026

Managing a LinkedIn Page means juggling post creation, comment moderation, media uploads, and analytics across multiple screens. Multiply that across several organization pages and the context-switching adds up fast.

The LinkedIn Community Management connector puts all of that behind 28 MCP tools so your Copilot Studio agent can handle it conversationally.

Seven operation categories

Posts (5 tools)

Full lifecycle management for LinkedIn posts:

The connector supports every LinkedIn post type: plain text, images, videos, articles with link previews, document carousel uploads, and reshares.

Comments (4 tools)

Reactions (3 tools)

Organizations (4 tools)

Social metadata (2 tools)

Media (7 tools)

Upload images, videos, and documents for use in posts:

Video upload flow:

Video files upload in 4 MB chunks:

1. Call initialize_video_upload with the file size to get chunk upload URLs and a video URN
2. PUT each 4 MB chunk to its corresponding upload URL and save the ETag from each response
3. Call finalize_video_upload with the video URN, upload token, and array of ETags

Statistics (3 tools)

Analytics for your LinkedIn Page:

Filter by time range using start_time and end_time as epoch milliseconds with optional granularity (DAY, WEEK, MONTH). Data is available for the past 12 months.

Example agent scenarios

Automated post creation and engagement monitoring

1. User tells the agent to draft a post about a new product launch
2. Agent uses create_post to publish text with an article link preview
3. Later, agent uses get_social_metadata to check likes, comments, and shares
4. Uses get_comments to surface any questions that need responses

Community engagement workflow

1. Agent uses find_posts_by_author to list recent organization posts
2. For each post, calls get_social_metadata to rank by engagement
3. Uses get_comments to find unanswered questions
4. Flags posts that need attention with comment counts and topics

Page analytics reporting

1. Agent uses get_follower_statistics to pull follower growth for the past quarter
2. Calls get_page_statistics to break down page views by audience demographics
3. Uses get_share_statistics to identify top-performing posts
4. Summarizes trends for the social media team

Rich media posting

1. Agent calls initialize_image_upload to get an upload URL
2. Uploads the image to the returned URL
3. Uses create_post with the image URN to publish a visual post
4. Tracks performance with get_social_metadata

LinkedIn API details

The connector handles several LinkedIn REST.li patterns automatically:

• URN encoding: Pass raw URN values like urn:li:organization:123. The connector URL-encodes them in API paths.
• Post creation response: LinkedIn returns the new post ID in the x-restli-id header. The connector extracts it and returns { "id": "urn:li:ugcPost:..." } in the response body.
• Partial updates: update_post and toggle_comments wrap your changes in the required {"patch":{"$set":{...}}} envelope automatically.
• Compound keys: delete_reaction constructs the REST.li compound key path /reactions/(actor:{urn},entity:{urn}) from individual parameters.
• API versioning: All requests include Linkedin-Version: 202602 and X-Restli-Protocol-Version: 2.0.0 headers automatically.

Prerequisites

• A LinkedIn account with admin access to at least one LinkedIn Page
• A LinkedIn Developer App with the Community Management product enabled
• OAuth 2.0 credentials (Client ID and Client Secret) from your LinkedIn Developer App

Required OAuth scopes

Setup

1. Go to the LinkedIn Developer Portal and sign in
2. Create a new app or select an existing one
3. Under Products, request access to Community Management
4. Under Auth, note your Client ID and Client Secret
5. Add the Power Platform redirect URL to Authorized redirect URLs

Deploy with PAC CLI:

pac connector create `
  --api-definition-file apiDefinition.swagger.json `
  --api-properties-file apiProperties.json `
  --script-file script.csx

For Copilot Studio:

1. Open your agent
2. Go to Actions > Add an action > Connector
3. Search for your connector and add it
4. The agent discovers all 28 tools automatically

Application Insights telemetry

Add your connection string to script.csx to track:

Resources

• GitHub repository
• LinkedIn Community Management API documentation
• LinkedIn Developer Portal