PostNitro Carousel Generator

Generate professional social media carousel posts using the PostNitro.ai Embed API. Supports AI-powered content generation and manual content import for Link...

Security Scan

The skill's requirements and runtime instructions match its stated purpose (calling the PostNitro Embed API to create carousels); nothing requested or described appears out of scope or unnecessary.

Purpose & Capability

Name/description describe creating social media carousels via PostNitro; required env vars (API key, template/brand/preset IDs) and the documented API endpoints align directly with that purpose. The primary credential (POSTNITRO_API_KEY) is appropriate and used as the auth header in all examples.

Instruction Scope

SKILL.md contains explicit curl examples: POST to initiate generate/import, GET to poll status and GET to retrieve output. The instructions only reference the PostNitro embed API, example public image URLs, and the declared environment variables. There are no steps that ask the agent to read unrelated files, secrets, or system state.

Install Mechanism

No install spec and no code files — instruction-only skill — so nothing is written to disk or fetched during install. This minimizes installation risk.

Credentials

Requires an API key plus three IDs used to select template/brand/preset. These are reasonable and proportionate for a hosted design-generation API. The skill does not request unrelated credentials or filesystem config paths.

Persistence & Privilege

always:false and no install actions mean the skill does not request elevated or permanent presence. It does not modify other skills or system configuration.

Assessment

This skill will send whatever you ask it to the PostNitro embed API (embed-api.postnitro.ai) using the API key you provide. Only grant an API key you trust PostNitro with, avoid sending sensitive/personal data through the skill, and prefer a scoped or revocable key if PostNitro supports it. Rotate or revoke the key if you stop using the skill, and check PostNitro's privacy/security docs if you need to know how uploaded content and generated images are stored or shared.

Generate social media carousel posts via the PostNitro.ai Embed API. Two workflows: AI generation (topic/article/X post → carousel) and content import (your own slides, with optional infographics).

Setup

1. Sign up at https://postnitro.ai (free plan: 5 credits/month)
2. Go to account settings → "Embed" → generate an API key
3. Set up your template, brand, and AI preset in the PostNitro dashboard
4. Set environment variables:

   export POSTNITRO_API_KEY="your-api-key"
   export POSTNITRO_TEMPLATE_ID="your-template-id"
   export POSTNITRO_BRAND_ID="your-brand-id"
   export POSTNITRO_PRESET_ID="your-ai-preset-id"
   

Base URL: https://embed-api.postnitro.ai Auth header: embed-api-key: $POSTNITRO_API_KEY

Core Workflow

All carousel creation is asynchronous: Initiate → Poll Status → Get Output.

1. Generate a carousel with AI

curl -X POST 'https://embed-api.postnitro.ai/post/initiate/generate' \
  -H 'Content-Type: application/json' \
  -H "embed-api-key: $POSTNITRO_API_KEY" \
  -d '{
    "postType": "CAROUSEL",
    "templateId": "'"$POSTNITRO_TEMPLATE_ID"'",
    "brandId": "'"$POSTNITRO_BRAND_ID"'",
    "presetId": "'"$POSTNITRO_PRESET_ID"'",
    "responseType": "PNG",
    "aiGeneration": {
      "type": "text",
      "context": "5 tips for growing your LinkedIn audience in 2026",
      "instructions": "Professional tone, actionable advice"
    }
  }'

Returns { "success": true, "data": { "embedPostId": "post123", "status": "PENDING" } }. Save the embedPostId.

aiGeneration.type values:

• "text" — context is the text content to turn into a carousel
• "article" — context is an article URL to extract and convert
• "x" — context is an X (Twitter) post or thread URL

See examples/generate-from-text.json, examples/generate-from-article.json, and examples/generate-from-x-post.json.

2. Import your own slide content

curl -X POST 'https://embed-api.postnitro.ai/post/initiate/import' \
  -H 'Content-Type: application/json' \
  -H "embed-api-key: $POSTNITRO_API_KEY" \
  -d '{
    "postType": "CAROUSEL",
    "templateId": "'"$POSTNITRO_TEMPLATE_ID"'",
    "brandId": "'"$POSTNITRO_BRAND_ID"'",
    "responseType": "PNG",
    "slides": [
      { "type": "starting_slide", "heading": "Your Title", "description": "Intro text" },
      { "type": "body_slide", "heading": "Key Point", "description": "Details here" },
      { "type": "ending_slide", "heading": "Take Action!", "cta_button": "Learn More" }
    ]
  }'

Returns same response format with embedPostId.

Slide rules:

• Exactly 1 starting_slide (required)
• At least 1 body_slide (required)
• Exactly 1 ending_slide (required)
• heading is required on every slide

Slide fields: heading (required), sub_heading, description, image (URL), background_image (URL), cta_button, layoutType, layoutConfig.

For infographic slides, set layoutType: "infographic" on body slides — replaces the image with structured data columns. See examples/import-infographics.json and references/api-reference.md for full infographics config.

3. Check post status

curl -X GET "https://embed-api.postnitro.ai/post/status/$EMBED_POST_ID" \
  -H "embed-api-key: $POSTNITRO_API_KEY"

Poll every 3–5 seconds until data.embedPost.status is "COMPLETED". The logs array shows step-by-step progress.

4. Get the output

curl -X GET "https://embed-api.postnitro.ai/post/output/$EMBED_POST_ID" \
  -H "embed-api-key: $POSTNITRO_API_KEY"

Returns downloadable URLs in data.result.data:

• PNG: Array of URLs (one per slide)
• PDF: Single URL

See references/api-reference.md for full response schemas.

Common Patterns

Pattern 1: LinkedIn thought leadership carousel

Generate a carousel from a topic with professional tone:

{
  "aiGeneration": {
    "type": "text",
    "context": "5 mistakes startups make with their LinkedIn strategy and how to fix each one",
    "instructions": "Professional but conversational tone. Each slide should have one clear takeaway."
  }
}

Pattern 2: Repurpose a blog post

Turn an existing article into a carousel:

{
  "aiGeneration": {
    "type": "article",
    "context": "https://yourblog.com/posts/social-media-strategy-2026",
    "instructions": "Extract the 5 most actionable points. Keep slide text concise."
  }
}

Pattern 3: Repurpose an X thread

Convert a viral X thread into a visual carousel:

{
  "aiGeneration": {
    "type": "x",
    "context": "https://x.com/username/status/1234567890",
    "instructions": "Maintain the original voice and key points"
  }
}

Pattern 4: Data-driven infographic carousel

Import slides with structured infographic layouts:

See examples/import-infographics.json for a complete example with grid and cycle layouts.

Content Strategy Tips

• LinkedIn: Professional tone, actionable insights, 6–10 slides, clear CTA.
• Instagram: Visual-first, concise text, 5–8 slides, storytelling arc.
• TikTok: Trendy, punchy, 4–7 slides, hook on slide 1.
• X (Twitter): Data-driven, 3–6 slides, provocative opening.

Common Gotchas

1. Default responseType is PDF — always specify "PNG" explicitly if you want individual slide images.
2. heading is required on every slide — omitting it returns an error.
3. Slide structure is strict — exactly 1 starting, at least 1 body, exactly 1 ending.
4. Article type needs a URL — "article" type expects a URL as context, not plain text.
5. X type needs an X post URL — "x" type expects https://x.com/... or https://twitter.com/... as context.
6. Infographics replace images — setting layoutType: "infographic" overrides any image on that slide.
7. Cyclical infographics use first column only — columnDisplay: "cycle" ignores data in columns 2+.
8. Max 3 columns — columnCount cannot exceed 3 for infographic layouts.
9. Image URLs must be public — image and background_image fields require publicly accessible URLs.
10. Credits vary by method — AI generation costs 2 credits/slide, content import costs 1 credit/slide.

Credits & Pricing

Plan	Price	Credits/Month
Free	$0	5
Monthly	$10	250+ (scalable)

• Content import: 1 credit per slide
• AI generation: 2 credits per slide

Supporting Resources

Reference docs:

• references/api-reference.md — Complete endpoint reference with request/response schemas and infographics config

Ready-to-use examples:

• examples/EXAMPLES.md — Index of all examples
• examples/generate-from-text.json — AI generation from text
• examples/generate-from-article.json — AI generation from article URL
• examples/generate-from-x-post.json — AI generation from X post
• examples/import-default.json — Basic slide import
• examples/import-infographics.json — Import with infographic layouts

Quick Reference

# Auth
Header: embed-api-key: $POSTNITRO_API_KEY

# AI generation
POST /post/initiate/generate  { postType, templateId, brandId, presetId, responseType?, requestorId?, aiGeneration: { type, context, instructions? } }

# Content import
POST /post/initiate/import  { postType, templateId, brandId, responseType?, requestorId?, slides: [{ type, heading, ... }] }

# Check status (poll until COMPLETED)
GET /post/status/{embedPostId}

# Get output (download URLs)
GET /post/output/{embedPostId}

Tips for the Agent

• Always confirm the user has set POSTNITRO_API_KEY, POSTNITRO_TEMPLATE_ID, POSTNITRO_BRAND_ID before calling any endpoint.
• POSTNITRO_PRESET_ID is only required for AI generation, not for content import.
• For "article" type, the context must be a URL — not article text. For article text, use "text" type.
• For "x" type, the context must be an X/Twitter post URL.
• Default responseType is "PDF" — always pass "PNG" if the user wants individual slide images.
• When importing slides, always structure as: 1 starting_slide → 1+ body_slide → 1 ending_slide.
• For data-heavy content, suggest using infographic layouts on body slides instead of plain text.
• Poll GET /post/status/{embedPostId} every 3–5 seconds — don't hammer the endpoint.
• After getting output, the data field contains download URLs — present them to the user directly.
• If the user doesn't specify a platform, suggest LinkedIn (most common carousel use case).
• Warn users about credit costs upfront: AI generation is 2x the cost of content import.