API Reference

Complete reference for the Skimming AI REST API

Overview

The Skimming AI API lets you build chat-with-your-content experiences programmatically. Upload files, ingest URLs, and chat with any content type.

---

Base URL

https://api.skimming.ai

For staging/testing:

https://staging-api.skimming.ai

---

Authentication

All requests require a Bearer token in the Authorization header.

Authorization: Bearer YOUR_API_KEY

Never expose your API key in client-side code. Use server-side requests only.

---

Required Headers

Every API request must include these headers:

Header	Value	Required
Authorization	Bearer YOUR_API_KEY	✅ Yes
Content-Type	application/json or multipart/form-data	✅ Yes

---

Core Workflow

1. Ingest

   Upload a file or URL to get a file_id.

   Endpoints:

   • POST /ingest/v1/api/document — Documents, images, audio, video
   • POST /ingest/v1/api/youtube — YouTube videos
   • POST /ingest/v1/api/website — Websites and social posts
2. Chat

   Use the file_id to ask questions about the content.

   Endpoints:

   • POST /chat/v1/api/document/{type} — PDF, DOCX, XLSX, PPTX, TXT, EPUB
   • POST /chat/v1/api/image/{type} — JPG, PNG, GIF, WebP, etc.
   • POST /chat/v1/api/video/{type} — YouTube, MP4, MOV, etc.
   • POST /chat/v1/api/audio/{type} — MP3, WAV, OGG, etc.
   • POST /chat/v1/api/website/{type} — Website, Instagram, X, LinkedIn, Facebook

---

Request Format

Ingest Request (File Upload)

curl -X POST https://api.skimming.ai/ingest/v1/api/document \
  -H "Authorization: Bearer sk-your-api-key" \
  -F "file=@document.pdf"

Chat Request (JSON)

curl -X POST https://api.skimming.ai/chat/v1/api/document/pdf \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "file_id": "abc123-def456-ghi789",
    "question": "What is this document about?",
    "streaming": false
  }'

---

Response Format

Success Response

{
  "success": true,
  "data": "The AI-generated answer...",
  "error": null,
  "message": {
    "content": null,
    "displayContent": true
  }
}

Error Response

{
  "success": false,
  "error": {},
  "message": {
    "content": "Error description",
    "displayContent": true
  }
}

---

Streaming

Set "streaming": true to receive responses as Server-Sent Events (SSE).

{
  "file_id": "abc123",
  "question": "Summarize this",
  "streaming": true
}

The response will stream in text/event-stream format.

---

Models

Specify the AI model with model and model_type:

Provider	model_type	Available Models
OpenAI	openai	gpt-4o-mini, gpt-4o, gpt-4, gpt-3.5-turbo
Anthropic	anthropic	claude-3-sonnet, claude-3.5-sonnet
Google	gemini	gemini-1.5-flash, gemini-1.5-pro

---

Rate Limits & Credits

Feature	Free Plan	Paid Plans
Credits	~1,000/month	Based on plan
File Size	5 MB	Up to 200 MB
Video/Audio	480 minutes	Based on plan

Credit usage per model:

• GPT-4o Mini / Gemini Flash: 1 credit
• GPT-4o: 5 credits
• Gemini Pro: 16 credits
• Claude 3.5/3.7: 25 credits
• GPT-4: 100 credits

---

HTTP Status Codes

Code	Description
200	Success
400	Bad request — invalid parameters
401	Unauthorized — invalid API key
402	Payment required — credits exceeded
500	Server error

---