Send a Notification

Pushmark is designed around a single HTTPS request to a public channel hash.

Endpoint

POST https://api.pushmark.app/<channel_hash>

Replace <channel_hash> with the hash shown in your channel page.

Payload shape

Public channel delivery is centered around a compact JSON body:

{
  "message": "Checkout API error rate passed 5% in the last 10 minutes.",
  "type": "error"
}

Integration examples

Use the tabs below for the fastest copy-paste starting points. If you need a language-specific variant, see More examples right after.

cURL
Pushmark CLI
JavaScript
Python
GitHub Actions

curl -X POST "https://api.pushmark.app/YOUR_CHANNEL_HASH" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "Nightly backup uploaded successfully.",
    "type": "success"
  }'

pushmark -t success YOUR_CHANNEL_HASH "Nightly backup uploaded successfully."

Install guide: pushmark/pushmark-cli

await fetch('https://api.pushmark.app/YOUR_CHANNEL_HASH', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    message: 'Deploy finished successfully.',
    type: 'success',
  }),
});

import requests

requests.post(
    "https://api.pushmark.app/YOUR_CHANNEL_HASH",
    json={
        "message": "CPU usage is back to normal.",
        "type": "info",
    },
)

- name: Notify Pushmark
  run: |
    curl -X POST "https://api.pushmark.app/${{ secrets.PUSHMARK_CHANNEL_HASH }}" \
      -H "Content-Type: application/json" \
      -d '{
        "message": "Release '${{ github.sha }}' is live.",
        "type": "success"
      }'

More examples

Go
Java
Swift
PHP cURL
PHP Guzzle

payload := map[string]string{
    "message": "Nightly backup uploaded successfully.",
    "type": "success",
}

jsonData, _ := json.Marshal(payload)
resp, _ := http.Post(
    "https://api.pushmark.app/YOUR_CHANNEL_HASH",
    "application/json",
    bytes.NewBuffer(jsonData),
)
defer resp.Body.Close()

HttpClient client = HttpClient.newHttpClient();
HttpRequest request = HttpRequest.newBuilder()
    .uri(URI.create("https://api.pushmark.app/YOUR_CHANNEL_HASH"))
    .header("Content-Type", "application/json")
    .POST(HttpRequest.BodyPublishers.ofString("{\"message\": \"Nightly backup uploaded successfully.\", \"type\": \"success\"}"))
    .build();

client.send(request, HttpResponse.BodyHandlers.ofString());

let url = URL(string: "https://api.pushmark.app/YOUR_CHANNEL_HASH")!
var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue("application/json", forHTTPHeaderField: "Content-Type")
request.httpBody = try! JSONSerialization.data(withJSONObject: [
    "message": "Nightly backup uploaded successfully.",
    "type": "success"
])

URLSession.shared.dataTask(with: request).resume()

<?php

$data = json_encode([
    "message" => "Nightly backup uploaded successfully.",
    "type" => "success",
]);

$ch = curl_init("https://api.pushmark.app/YOUR_CHANNEL_HASH");
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);
curl_setopt($ch, CURLOPT_HTTPHEADER, ["Content-Type: application/json"]);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);

$response = curl_exec($ch);
curl_close($ch);

<?php

use GuzzleHttp\Client;

$client = new Client();
$response = $client->post("https://api.pushmark.app/YOUR_CHANNEL_HASH", [
    "json" => [
        "message" => "Nightly backup uploaded successfully.",
        "type" => "success",
    ],
]);

echo $response->getBody();

Notification types

Current product examples use a small set of notification types:

info
success
warning
error
log

Delivery expectations

Send from a trusted server, automation, or CI environment
Keep the message concise enough to scan at a glance
Use separate public channels for unrelated workflows
Treat private channels as coming soon

Using Pushmark from AI agents

If you want an AI agent to work with Pushmark directly, use the Pushmark MCP endpoint instead of asking the model to generate raw HTTP requests every time.

That setup works well with tools like Codex, Cursor, Claude Code, and other MCP-compatible clients. See AI Agents and MCP for: