API Overview

The ConnectSafely API surfaces LinkedIn automation features that power the product UI. Each endpoint documented here comes directly from the main app and supports live testing via the Try it now playground.

Authentication

• Generate an API key from connectsafely.ai/api-key.
• Send it with every authenticated request as a Bearer token.
• Rotate keys regularly and never expose production keys in client-side code.

Authorization: Bearer <YOUR_API_KEY>

Environments

Configure the docs playground using the Docs environment (proxied through this site). For production traffic, call the API from your backend services directly.

Endpoint Groups

• LinkedIn Account – Manage account status, activity history, and warmup settings.
• LinkedIn Actions – Follow, connect, and manage relationships with members.
• LinkedIn Posts – Read, react, comment, and create posts.
• LinkedIn Profiles – Retrieve enriched profile data and visitors.
• LinkedIn Groups – Search and manage LinkedIn groups.
• LinkedIn Messaging – Send messages and manage conversations.
• LinkedIn Search – Search for people, companies, jobs, and more.
• LinkedIn Analytics – Access creator analytics and engagement data.

Use the sidebar navigation to browse each group or jump directly using the shortcuts below:

• LinkedIn Account
• LinkedIn Actions
• LinkedIn Posts
• LinkedIn Profiles
• LinkedIn Groups
• LinkedIn Messaging
• LinkedIn Search
• LinkedIn Analytics

Rate Limits

All limits are per LinkedIn account:

• Spacing: calls to the same account are released about 3 seconds apart. Calls to different accounts run in parallel.
• Per-minute limits: 15 per minute for actions (connect, message, comment, react, follow), 30 per minute for searches and reads, 10 per minute for account calls. Inbox reads have no per-minute limit.
• Daily and weekly limits: for example 90 connection requests per week and 150 opening messages per day. You can change these per account.

When a limit is reached the API returns 429. See Rate Limits for every limit, what counts against it, the response headers and how to handle each error.

Error Handling

Errors follow a consistent shape:

{
  "success": false,
  "code": "validation_error",
  "message": "Explanation of what went wrong"
}

Always check the HTTP status code and the success field before processing data.

Changelog

Endpoint docs are generated from the canonical specification inside the web application. Regenerate the docs (bun run generate:api) whenever the product team updates the in-app documentation.