Preview a comment prompt against a sample post

curl --request POST \
  --url https://api.reply.io/v3/ai-prompts/preview/sample \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "promptText": "Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.",
  "skipTopics": [
    "politics",
    "religion",
    "personal topics"
  ],
  "sequenceId": 4521
}
'

import requests

url = "https://api.reply.io/v3/ai-prompts/preview/sample"

payload = {
    "promptText": "Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.",
    "skipTopics": ["politics", "religion", "personal topics"],
    "sequenceId": 4521
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    promptText: 'Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.',
    skipTopics: ['politics', 'religion', 'personal topics'],
    sequenceId: 4521
  })
};

fetch('https://api.reply.io/v3/ai-prompts/preview/sample', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.reply.io/v3/ai-prompts/preview/sample",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'promptText' => 'Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.',
    'skipTopics' => [
        'politics',
        'religion',
        'personal topics'
    ],
    'sequenceId' => 4521
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.reply.io/v3/ai-prompts/preview/sample"

	payload := strings.NewReader("{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.reply.io/v3/ai-prompts/preview/sample")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.reply.io/v3/ai-prompts/preview/sample")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}"

response = http.request(request)
puts response.read_body

{
  "outcome": "generated",
  "commentText": "The point about onboarding friction really lands — we saw the same drop-off until we cut the setup steps in half.",
  "postText": "After six months of user interviews, the biggest blocker we found wasn't pricing. It was onboarding friction.",
  "detectedTopic": "product onboarding",
  "postUrl": "https://www.linkedin.com/feed/update/urn:li:activity:7100000000000000000"
}

AI Prompts

Preview a comment prompt against a sample post

Coming soon. This endpoint will be available by late August 2026.

Requires the sequences:operate scope (or a broader one that includes it).

Use this endpoint to try out a comment prompt against a generic sample post while you are still writing it. Nothing is fetched from LinkedIn and no Reply credits are spent, so this is the cheap way to iterate on wording. Pass sequenceId if the prompt text contains sequence variables and you want them filled in for the preview. Check outcome to see whether a comment was produced: generated means commentText holds the result, while any other value explains why the AI produced nothing. The number of sample previews you can run per day is capped.

POST

ai-prompts

preview

sample

Preview a comment prompt against a sample post

curl --request POST \
  --url https://api.reply.io/v3/ai-prompts/preview/sample \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "promptText": "Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.",
  "skipTopics": [
    "politics",
    "religion",
    "personal topics"
  ],
  "sequenceId": 4521
}
'

import requests

url = "https://api.reply.io/v3/ai-prompts/preview/sample"

payload = {
    "promptText": "Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.",
    "skipTopics": ["politics", "religion", "personal topics"],
    "sequenceId": 4521
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    promptText: 'Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.',
    skipTopics: ['politics', 'religion', 'personal topics'],
    sequenceId: 4521
  })
};

fetch('https://api.reply.io/v3/ai-prompts/preview/sample', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.reply.io/v3/ai-prompts/preview/sample",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'promptText' => 'Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.',
    'skipTopics' => [
        'politics',
        'religion',
        'personal topics'
    ],
    'sequenceId' => 4521
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.reply.io/v3/ai-prompts/preview/sample"

	payload := strings.NewReader("{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.reply.io/v3/ai-prompts/preview/sample")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.reply.io/v3/ai-prompts/preview/sample")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"promptText\": \"Write a short, genuine comment that reacts to one concrete point in the post. Avoid flattery and never pitch.\",\n  \"skipTopics\": [\n    \"politics\",\n    \"religion\",\n    \"personal topics\"\n  ],\n  \"sequenceId\": 4521\n}"

response = http.request(request)
puts response.read_body

{
  "outcome": "generated",
  "commentText": "The point about onboarding friction really lands — we saw the same drop-off until we cut the setup steps in half.",
  "postText": "After six months of user interviews, the biggest blocker we found wasn't pricing. It was onboarding friction.",
  "detectedTopic": "product onboarding",
  "postUrl": "https://www.linkedin.com/feed/update/urn:li:activity:7100000000000000000"
}

Authorizations

Authorization

string

header

required

Authenticate every request with a Bearer token. Pass your Reply API key in the Authorization header:

Authorization: Bearer <your-api-key>

Get your API key from the Reply dashboard: Settings → API Key.

Body

application/json

Request body for previewing a comment prompt against a generic sample post.

promptText

string

required

The prompt instructions to test

Maximum string length: 2000

skipTopics

string[]

Topics the AI should refuse to comment on. A post matching one of these returns the aiSkip outcome.

sequenceId

integer | null

Sequence to resolve variables against when promptText contains sequence variables. Omit if the prompt has no variables.

Required range: x >= 1

Response

The preview result

The result of a comment preview. Read outcome first — it tells you whether a comment was produced and, if not, why.

outcome

enum<string>

required

Whether a comment was generated, and why not if it was skipped

Available options:

generated,

noRecentPosts,

postsUnavailable,

commentingRestricted,

reshareNoText,

noPostContent,

aiSkip

commentText

string | null

The generated comment. Populated only when outcome is generated.

postText

string | null

The text of the post the comment was written for

detectedTopic

string | null

The topic the AI detected in the post. Useful for checking why a post was skipped.

postUrl

string | null

Link to the post the comment was written for

Getting started

For AI agents

API Documentation

Schemas

Webhooks

Machine-readable

Preview a comment prompt against a sample post

Authorizations

Body

Response