> ## Documentation Index > Fetch the complete documentation index at: https://haider-5d8ca6e7.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # Chat with an Instagram post > Ask questions about an Instagram post using its `file_id`. **Workflow:** 1. Ingest Instagram URL using `/source/v1/api/website` with JSON body containing `url` 2. Use returned `file_id` with this endpoint ## OpenAPI ````yaml openapi.yaml post /chat/v1/api/website/instagram openapi: 3.0.3 info: title: Skimming AI API version: 1.0.0 description: > Build **PDF.ai / Chatbase-style** "chat with your content" experiences using a simple REST API. ## What you can do - Chat with PDFs, documents, spreadsheets, slides, images, audio, video, websites, and social posts - Parse documents into clean structured content - Extract transcripts/content from media and websites ## Authentication All requests require a Bearer token. **Header** `Authorization: Bearer YOUR_API_KEY` ## Base URL Use one of the servers below. ## Important Notes - **file_id is required**: You must first upload/source a file using `/source/v1/api/document` or `/source/v1/api/youtube` to get a `file_id` - **Streaming**: Chat endpoints support streaming responses (SSE format) when `streaming: true` ## Quick Start Guide **1. Get your API Key** - Generate an API key from your account dashboard - Use format: `Authorization: Bearer sk-your-api-key-here` **2. Ingest a file** ``` POST /source/v1/api/document Authorization: Bearer sk-your-api-key Content-Type: multipart/form-data source: [your-file.pdf] ``` Response contains `file_id` - save this! **3. Chat with the file** ``` POST /chat/v1/api/document/pdf Authorization: Bearer sk-your-api-key Content-Type: application/json { "file_id": "your-file-id-from-step-2", "question": "What is this document about?", "streaming": false } ``` **Required Headers for all requests:** - `Authorization: Bearer YOUR_API_KEY` (required) **Notes:** - Do **not** send `platform`, `userid`, `useremail` for API calls. The backend derives the user from your API key. contact: name: Skimming AI Support url: https://www.skimming.ai email: support@skimming.ai license: name: MIT url: https://opensource.org/licenses/MIT servers: - url: https://app.skimming.ai/back description: Production - url: https://staging-app.skimming.ai/back description: Staging security: - bearerAuth: [] tags: - name: Ingest description: > Upload and ingest files to get a `file_id` for chat operations. **Endpoints:** - `/source/v1/api/document` - Ingest documents (PDF, , CSV, EPUB, PPTX, TXT) - `/source/v1/api/image` - Ingest images (JPG, PNG, GIF, WebP, etc.) - `/source/v1/api/audio` - Ingest audio files (MP3, WAV, OGG, etc.) - `/source/v1/api/video` - Ingest video files (MP4, MOV, AVI, etc.) - `/source/v1/api/youtube` - Ingest YouTube videos - `/source/v1/api/website` - Ingest websites and social media posts **Returns:** `file_id` which is required for all chat operations. - name: Document description: | Chat with document files (PDF, Excel, PowerPoint, Text, EPUB). **Workflow:** 1. Ingest document using `/source/v1/api/document` 2. Get `file_id` from response 3. Use appropriate `/chat/v1/api/document/:type` endpoint **Available Types:** - `pdf` - PDF files - `xlsx` - Excel/CSV files (.xlsx, .xls, .csv) - `pptx` - PowerPoint presentations (.pptx, .ppt) - `txt` - Text files (.txt, .rtf) - `epub` - EPUB files - name: Image description: | Chat with image files using AI-powered vision. **Workflow:** 1. Ingest image using `/source/v1/api/image` 2. Get `file_id` from response 3. Use `/chat/v1/api/image/:type` endpoint **Available Types:** - `jpg` / `jpeg` - JPEG images - `png` - PNG images - `gif` - GIF images - `webp` - WebP images - `bmp` - BMP images - `tiff` - TIFF images - `svg` - SVG images - `heic` - HEIC images - name: Video description: | Chat with video content including YouTube videos. **Workflow for YouTube:** 1. Ingest YouTube video using `/source/v1/api/youtube` 2. Get `file_id` from response 3. Use `/chat/v1/api/video/youtube` endpoint **Workflow for other videos:** 1. Ingest video file using `/source/v1/api/video` 2. Get `file_id` from response 3. Use `/chat/v1/api/video/:type` endpoint **Available Types:** - `youtube` - YouTube videos - `mp4` - MP4 videos - `mov` - MOV videos - `avi` - AVI videos - `wmv` - WMV videos - `webm` - WebM videos - `flv` - FLV videos - `mkv` - MKV videos - name: Audio description: | Chat with audio files and transcripts. **Workflow:** 1. Ingest audio using `/source/v1/api/audio` 2. Get `file_id` from response 3. Use `/chat/v1/api/audio/:type` endpoint **Available Types:** - `mp3` - MP3 audio - `wav` - WAV audio - `ogg` - OGG audio - `aac` - AAC audio - `flac` - FLAC audio - `m4a` - M4A audio - `webm` - WebM audio - name: Source description: | Chat with websites and social media posts. **Workflow:** 1. Ingest URL using `/source/v1/api/website` 2. Get `file_id` from response 3. Use appropriate `/chat/v1/api/website/:type` endpoint **Available Types:** - `website` - General websites - `instagram` - Instagram posts - `twitter` - Twitter/X posts - `linkedin` - LinkedIn posts - `facebook` - Facebook posts - name: Parse description: | Parse content (no DB save). Returns extracted text + metadata. **Supported types:** - `document`, `image`, `video`, `audio` (multipart upload `source`) - `website`, `youtube` (JSON body `{ "source": "..." }`) - name: Extract description: | Extract text only (no DB save). Returns `textContent`. **Supported types:** - `document`, `image`, `video`, `audio` (multipart upload `source`) - `website`, `youtube` (JSON body `{ "source": "..." }`) - name: Usage description: > API usage and API plan totals for frontend dashboards. **Endpoints:** - `/users/api-count` - API monthly usage (counts) - `/users/api-plan-summary` - API plan totals only (questions/docs/hours per month) paths: /chat/v1/api/website/instagram: post: tags: - Source summary: Chat with an Instagram post description: > Ask questions about an Instagram post using its `file_id`. **Workflow:** 1. Ingest Instagram URL using `/source/v1/api/website` with JSON body containing `url` 2. Use returned `file_id` with this endpoint operationId: chatWithInstagram requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ChatRequest' examples: example: summary: Chat with Instagram post value: file_id: abc123-def456-ghi789 question: What is this post about? streaming: false responses: '200': $ref: '#/components/responses/ChatOK' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '402': description: Question limit exceeded content: application/json: schema: $ref: '#/components/schemas/Error' components: schemas: ChatRequest: type: object additionalProperties: false required: - file_id - question properties: file_id: type: string description: The file_id returned from ingest endpoint example: abc123-def456-ghi789 question: type: string description: Your question about the content example: What are the main points discussed in this document? streaming: type: boolean default: true description: Whether to stream the response (SSE format) model: type: string default: gpt-4 description: AI model to use enum: - gpt-4 - gpt-4o - gpt-4o-mini - gpt-3.5-turbo - claude-3-opus - claude-3-sonnet - claude-3.5-sonnet - gemini-1.5-pro - gemini-1.5-flash example: gpt-4 model_type: type: string default: openai description: Model provider enum: - openai - anthropic - gemini example: openai conversation_type: type: string enum: - single - group default: single description: Type of conversation max_output: type: integer description: Maximum output tokens example: 2000 Error: type: object properties: success: type: boolean default: false error: type: object description: Error details message: type: object properties: content: type: string description: Error message example: Question limit exceeded! displayContent: type: boolean ChatResponse: type: object properties: success: oneOf: - type: boolean - type: string description: Response content or success status data: type: string description: The AI-generated answer (when non-streaming) example: The main points discussed in this content are... error: type: object nullable: true description: Error details if any message: type: object properties: content: type: string nullable: true displayContent: type: boolean responses: ChatOK: description: Successful response content: text/event-stream: schema: type: string description: Server-Sent Events stream (when streaming=true) application/json: schema: $ref: '#/components/schemas/ChatResponse' examples: non_streaming: summary: Non-streaming response value: success: true data: The main points discussed in this content are... error: null message: content: null displayContent: true BadRequest: description: Bad request - invalid parameters content: application/json: schema: $ref: '#/components/schemas/Error' example: success: false error: {} message: content: Invalid request parameters displayContent: true Unauthorized: description: Unauthorized - invalid or missing API key content: application/json: schema: $ref: '#/components/schemas/Error' example: success: false error: {} message: content: Invalid API key displayContent: true securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: APIKey description: Bearer token authentication using an API key ````