Developer API · v1

Your LinkedIn, via API,
now programmable.

A clean REST interface across the whole content pipeline: create, schedule and publish posts, read analytics, receive webhooks. JSON in, JSON out.

Open the interactive reference Quickstart →

Base https://app.tidefall.de/api/v1

API & MCP access is included in every plan.

Isolated per profile

An API key belongs to exactly one LinkedIn profile. It sees and changes that profile’s data only — never anyone else’s.

REST · JSON · versioned

Predictable resources under /api/v1, cursor pagination, a uniform error envelope and idempotency keys.

Async, honestly

Long-running steps (image, publish) answer with 202 — you poll the status or receive a signed webhook.

Quickstart

Your first call in two minutes.

1

Create a key

In the dashboard under Settings → Integrations, create an API key. It is shown only once — store it safely.
2

Authenticate

Send every request with the header Authorization: Bearer lps_…
3

Get going

Use GET /me to check which profile your key is connected to — then posts, ideas, analytics.

request

curl https://app.tidefall.de/api/v1/me \
  -H "Authorization: Bearer lps_your_key"

200 · response

{
  "member": {
    "id": "371707cb-…",
    "display_name": "Jane Doe",
    "linkedin_connected": true
  }
}

Reference

All endpoints at a glance.

Grouped by resource. For schemas, examples and “Try it”, head to the interactive reference.

Posts

12 endpoints

The complete lifecycle — from draft to publication on LinkedIn.

POST /posts Create a post draft from an idea
GET /posts List posts (cursor pagination)
GET /posts/{id} Get a post
PATCH /posts/{id} Edit the post text
DELETE /posts/{id} Delete an unpublished post
POST /posts/{id}/regenerate Rewrite (different template)
POST /posts/{id}/images 202 Generate image candidates (AI)
POST /posts/{id}/images/upload Upload your own image (file or URL)
POST /posts/{id}/select-image Select an image
POST /posts/{id}/schedule Schedule for a point in time
POST /posts/{id}/publish 202 Publish on LinkedIn
GET /posts/{id}/analytics Engagement metrics for a post

Ideas

7 endpoints

Find content ideas, research them and turn them into drafts.

GET /ideas List ideas (new · used · dismissed)
POST /ideas/generate Generate ideas from your sources
POST /ideas/research 202 Research a keyword
POST /ideas/{id}/create-post Idea → post draft
PATCH /ideas/{id} Edit title / angle
POST /ideas/{id}/dismiss Dismiss an idea
POST /ideas/{id}/restore Bring back a dismissed idea

Analytics & account

3 endpoints

Measure performance and check which profile your key is connected to.

GET /analytics Account KPIs + top posts
GET /me Your connected profile
GET /health API status (public, no key)

Webhooks

5 endpoints

Get notified instead of polling — signed and with retries.

GET /webhooks List registered endpoints
POST /webhooks Register an endpoint (secret shown once)
GET /webhooks/{id} Get an endpoint
DELETE /webhooks/{id} Remove an endpoint
GET /webhooks/{id}/deliveries Inspect delivery attempts

MCP

Or just talk to Claude.

The same key also works as an MCP server for Claude Desktop, Cursor & co. — research, write a post, create an image/video/carousel, schedule or publish, right from the chat.

1

Create a key

In the dashboard under Account → Developer, create a key.
2

Configure the client

Add it as an MCP server in Claude Desktop / Cursor (see the config on the right).
3

Restart

All tools appear in the client’s tool menu — the list there is always up to date.

Content ideas · research · post drafts · images/video/carousels · scheduling & publishing · analytics — the full, always up-to-date tool list is in the MCP reference.

claude_desktop_config.json

{
  "mcpServers": {
    "linkedin-posting-system": {
      "url": "https://app.tidefall.de/api/mcp",
      "headers": { "Authorization": "Bearer lps_your_key" }
    }
  }
}

Async & webhooks

First the job, then the signal.

Image generation and publishing run in the background. The call replies immediately with 202 Accepted. You then poll GET /posts/{id} — or have a webhook notify you.

post.published A post was published on LinkedIn.
post.publish_failed A publication failed for good.
post.images_ready Image candidates have finished generating.
idea.created New ideas were created.

Every delivery carries an HMAC signature over the raw body — that’s how you verify it really comes from us:

POST → your endpoint · headers

X-LPS-Event: post.published
X-LPS-Delivery-Id: 9f2c…
X-LPS-Signature: sha256=<hmac(secret, rawBody)>

Verification · Node

const sig = "sha256=" + crypto
  .createHmac("sha256", secret)
  .update(rawBody)
  .digest("hex");
// compare timing-safe with X-LPS-Signature

Errors

One error, one format.

Every error arrives in the same envelope — the code is the contract, the message is a hint.

422 · response

{
  "error": {
    "code": "validation_error",
    "message": "Invalid request body",
    "details": { "idea": "required" }
  }
}

Rate limits

Per key, per endpoint bucket. Every response carries X-RateLimit-Limit, -Remaining and -Reset. When exceeded: 429.

Code	HTTP	When
`unauthorized`	401	Key is missing, invalid or revoked.
`not_found`	404	Object does not exist — or does not belong to your profile.
`validation_error`	422	Body/query invalid (details per field).
`conflict`	409	State does not allow it (e.g. already published).
`rate_limited`	429	Key limit reached — try again later.
`internal_error`	500	Unexpected server error.

Ready to build?

Open the interactive reference and try every endpoint right in the browser — or grab the machine-readable spec as OpenAPI 3.1.

Interactive reference OpenAPI 3.1 ↗

Your LinkedIn, via API, now programmable.