Summarize

https://summarize.sh/docs/
Admin
Summarize is a tool that processes content from links, websites, and YouTube videos to extract clean...
Tokens:14,876
Snippets:110
Trust Score:2.6
Update:8 months ago
Show doc for...
Tokens:
Raw
### Summarize Daemon Installation

Source: https://summarize.sh/docs/chrome-extension

Instructions for installing the Summarize daemon service on various operating systems.

```APIDOC
## Summarize Daemon Installation

### Description
Installs the Summarize daemon as a background service.

### Method
CLI Command

### Endpoint
N/A

### Parameters
#### CLI Arguments
- **--token** (string) - Required - Authentication token for the daemon.
- **--dev** (boolean) - Optional - Installs in development mode, running directly from source.

### Request Example
```bash
summarize daemon install --token <YOUR_TOKEN>
```

### Response
#### Success Response
Daemon installed and running.

#### Response Example
```
Daemon installed successfully.
```
```

--------------------------------

### Install Summarize CLI Tool

Source: https://summarize.sh/docs/chrome-extension

Installs the Summarize command-line interface globally using npm. This is the first step for setting up the Summarize extension and daemon.

```bash
npm i -g @steipete/summarize
```

--------------------------------

### Start Development Server for Extension

Source: https://summarize.sh/docs/chrome-extension

Starts the development server for the Chrome extension. This allows for live reloading and debugging during development.

```bash
pnpm -C apps/chrome-extension dev
```

--------------------------------

### Install Summarize CLI Tool (macOS ARM64)

Source: https://summarize.sh/docs/chrome-extension

Installs the Summarize command-line interface for macOS on ARM64 architecture using Homebrew. This provides an alternative installation method.

```bash
brew install steipete/tap/summarize
```

--------------------------------

### Install Summarize Daemon Service

Source: https://summarize.sh/docs/chrome-extension

Installs the Summarize daemon as a background service. This command requires a token obtained from the extension and sets up the service for macOS, Linux, or Windows.

```bash
summarize daemon install --token <TOKEN>
```

--------------------------------

### Install Playwright Browsers

Source: https://summarize.sh/docs/chrome-extension

Installs the necessary browser binaries for Playwright to run its tests. This is typically needed only on the first run of the E2E tests.

```bash
pnpm -C apps/chrome-extension exec playwright install chromium
```

--------------------------------

### Install Summarize Daemon Service (Development)

Source: https://summarize.sh/docs/chrome-extension

Installs the Summarize daemon in development mode. This allows the service to run directly from source code without needing a build, useful during development.

```bash
pnpm summarize daemon install --token <TOKEN> --dev
```

--------------------------------

### GET /v1/ping

Source: https://summarize.sh/docs/chrome-extension

A simple endpoint to check if the daemon is responsive. Requires authentication.

```APIDOC
## GET /v1/ping

### Description
A simple endpoint to check if the daemon is responsive. Requires authentication.

### Method
GET

### Endpoint
/v1/ping

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Response
#### Success Response (200)
- **ok** (boolean) - Indicates a successful ping.

#### Response Example
```json
{
  "ok": true
}
```
```

--------------------------------

### JSON Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Example of a JSON configuration file for the summarize_sh project. This file allows users to set default values for various options, including the model, environment variables, output language, prompt, and UI theme. It demonstrates both detailed and shorthand notations for certain settings.

```json
{
  "model": { "id": "google/gemini-3-flash-preview" },
  "env": { "OPENAI_API_KEY": "sk-..." },
  "output": { "language": "auto" },
  "prompt": "Explain like I am five.",
  "ui": { "theme": "ember" }
}

```

```json
{
  "model": "google/gemini-3-flash-preview"
}

```

```json
{
  "model": { "mode": "auto" }
}

```

```json
{
  "model": "auto"
}

```

--------------------------------

### Summarize SSE Stream Endpoint (GET)

Source: https://summarize.sh/docs/chrome-extension

The Server-Sent Events (SSE) endpoint on the daemon that streams the summarized content. The panel page connects to this endpoint to render the Markdown.

```javascript
GET /v1/summarize/<id>/events
```

--------------------------------

### GET /health

Source: https://summarize.sh/docs/chrome-extension

Checks the health status of the daemon. Returns a simple JSON object indicating if the daemon is operational.

```APIDOC
## GET /health

### Description
Checks the health status of the daemon. Returns a simple JSON object indicating if the daemon is operational.

### Method
GET

### Endpoint
/health

### Response
#### Success Response (200)
- **ok** (boolean) - Indicates if the daemon is healthy.
- **pid** (integer) - The process ID of the daemon.

#### Response Example
```json
{
  "ok": true,
  "pid": 12345
}
```
```

--------------------------------

### GET /v1/summarize/<id>/slides/events

Source: https://summarize.sh/docs/chrome-extension

Streams slide updates related to a specific summarization job using Server-Sent Events (SSE).

```APIDOC
## GET /v1/summarize/<id>/slides/events

### Description
Streams slide updates related to a specific summarization job using Server-Sent Events (SSE). This stream provides updates on slide content and status changes independent of the main summarization stream.

### Method
GET

### Endpoint
/v1/summarize/<id>/slides/events

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the summarization job.

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Response
#### SSE Stream Events
- **event: slides** - Data containing updated slide information.
- **event: status** - Data indicating the current status of the slide generation.
- **event: done** - Event signaling that all slide updates are complete.
- **event: error** - Event containing error details if an issue occurred during slide generation.
```

--------------------------------

### GET /v1/summarize/:id/events

Source: https://summarize.sh/docs/chrome-extension

Streams events related to a summarization job using Server-Sent Events (SSE). Provides real-time updates on progress, metrics, and results.

```APIDOC
## GET /v1/summarize/:id/events

### Description
Streams events related to a summarization job using Server-Sent Events (SSE). Provides real-time updates on progress, metrics, and results.

### Method
GET

### Endpoint
/v1/summarize/:id/events

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the summarization job.

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Response
#### SSE Stream Events
- **event: chunk** - Data containing text chunks of the summary (`data: { "text": "..." }`).
- **event: meta** - Metadata about the summarization, including the model used (`data: { "model": "..." }`).
- **event: status** - Progress messages displayed before the summary output begins (`data: { "text": "Processing..." }`).
- **event: metrics** - Performance metrics for the summarization job (`data: { "elapsedMs": 1234, "summary": "...", "details": "...", "summaryDetailed": "...", "detailsDetailed": "..." }`).
- **event: done** - Event signaling the completion of the summarization job (`data: {}`).
- **event: error** - Event containing error details if an issue occurred during summarization (`data: { "message": "Error details" }`).
```

--------------------------------

### Define Model Presets for Summarize SH

Source: https://summarize.sh/docs/config

Configure custom model presets that can be selected using the `--model <preset>` command-line argument. This allows for easy switching between different model configurations, including free tier options.

```json
{
  "models": {
    "fast": { "id": "openai/gpt-5-mini" },
    "or-free": {
      "rules": [
        {
          "candidates": [
            "openrouter/google/gemini-2.0-flash-exp:free",
            "openrouter/meta-llama/llama-3.3-70b-instruct:free"
          ]
        }
      ]
    }
  }
}
```

--------------------------------

### Summarize Daemon Management

Source: https://summarize.sh/docs/chrome-extension

Commands for managing the Summarize daemon service.

```APIDOC
## Summarize Daemon Management

### Description
Commands to check the status, start, and stop the Summarize daemon.

### Method
CLI Command

### Endpoint
N/A

### Parameters
None

### Request Example
```bash
summarize daemon status
summarize daemon restart
```

### Response
#### Success Response
Daemon status information or service restarted.

#### Response Example
```
Daemon is running.
```
```

--------------------------------

### Configure CLI Behavior and Fallbacks for Summarize SH

Source: https://summarize.sh/docs/config

Customize the behavior of the command-line interface, including enabling specific providers, configuring auto-fallback mechanisms, and setting provider-specific binary paths and arguments.

```json
{
  "cli": {
    "enabled": ["gemini", "agent"],
    "autoFallback": {
      "enabled": true,
      "onlyWhenNoApiKeys": true,
      "order": ["claude", "gemini", "codex", "agent"]
    },
    "codex": { "model": "gpt-5.2" },
    "claude": { "binary": "/usr/local/bin/claude", "extraArgs": ["--verbose"] },
    "agent": { "binary": "/usr/local/bin/agent", "model": "gpt-5.2" }
  }
}
```

--------------------------------

### Configure OpenAI API Endpoint and Settings

Source: https://summarize.sh/docs/config

Set a custom base URL for the OpenAI API, enable chat completions, and specify Whisper transcription costs per minute. This is useful for using proxies or compatible API endpoints.

```json
{
  "openai": {
    "baseUrl": "https://my-openai-proxy.example.com/v1",
    "useChatCompletions": true,
    "whisperUsdPerMinute": 0.006
  }
}
```

--------------------------------

### UI Theme Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Configuration snippet for setting the default UI theme in summarize_sh. This allows users to specify a preferred theme for the command-line interface.

```json
{
  "ui": { "theme": "moss" }
}

```

--------------------------------

### Prompt Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Configuration snippet for setting a custom prompt in summarize_sh. This allows overriding the built-in summary instructions with user-defined prompts, affecting how the content is summarized.

```json
{
  "prompt": "Explain for a kid. Short sentences. Simple words."
}

```

--------------------------------

### Slides Defaults Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Configuration for enabling slides generation by default in summarize_sh and tuning its parameters. This includes settings for OCR, output directory, scene detection thresholds, and maximum/minimum durations for slides.

```json
{
  "slides": {
    "enabled": true,
    "ocr": false,
    "dir": "slides",
    "sceneThreshold": 0.3,
    "max": 20,
    "minDuration": 2
  }
}

```

--------------------------------

### Build Firefox Extension

Source: https://summarize.sh/docs/chrome-extension

Builds the Firefox version of the Chrome extension. The output can be loaded as a temporary add-on via Firefox's about:debugging page.

```bash
pnpm -C apps/chrome-extension build:firefox
```

--------------------------------

### Logging Configuration for Summarize Sh Daemon

Source: https://summarize.sh/docs/config

Configuration for enabling and customizing JSON logging for the summarize_sh daemon. This includes settings for log level, format, file path, and rotation based on file size and number of files.

```json
{
  "logging": {
    "enabled": true,
    "level": "info",
    "format": "json",
    "file": "~/.summarize/logs/daemon.jsonl",
    "maxMb": 10,
    "maxFiles": 3
  }
}

```

--------------------------------

### Override Provider API Base URLs

Source: https://summarize.sh/docs/config

Configure custom base URLs for various AI providers to route requests through proxies, gateways, or compatible APIs. Environment variables offer an alternative and take precedence.

```json
{
  "openai": { "baseUrl": "https://my-openai-proxy.example.com/v1" },
  "nvidia": { "baseUrl": "https://integrate.api.nvidia.com/v1" },
  "anthropic": { "baseUrl": "https://my-anthropic-proxy.example.com" },
  "google": { "baseUrl": "https://my-google-proxy.example.com" },
  "xai": { "baseUrl": "https://my-xai-proxy.example.com" }
}
```

--------------------------------

### Run End-to-End Tests (Chrome Extension)

Source: https://summarize.sh/docs/chrome-extension

Executes the end-to-end tests for the Chrome extension using Playwright. This command helps ensure the extension functions as expected.

```bash
pnpm -C apps/chrome-extension test:e2e
```

--------------------------------

### Configure Auto Model Selection with Rules in Summarize SH

Source: https://summarize.sh/docs/config

Enable automatic model selection based on content type and token limits using a set of defined rules. This provides dynamic model switching for optimal performance and cost-efficiency.

```json
{
  "model": {
    "mode": "auto",
    "rules": [
      {
        "when": ["video"],
        "candidates": ["google/gemini-3-flash-preview"]
      },
      {
        "when": ["website", "youtube"],
        "bands": [
          {
            "token": { "max": 8000 },
            "candidates": ["openai/gpt-5-mini"]
          },
          {
            "candidates": ["xai/grok-4-fast-non-reasoning"]
          }
        ]
      },
      {
        "candidates": ["openai/gpt-5-mini", "openrouter/openai/gpt-5-mini"]
      }
    ]
  },
  "media": { "videoMode": "auto" }
}
```

--------------------------------

### Verify Daemon Connection (cURL)

Source: https://summarize.sh/docs/chrome-extension

Sends a health check request to the Summarize daemon's HTTP server using cURL. This verifies that the daemon is running and accessible at its local address.

```bash
curl http://127.0.0.1:8787/health
```

--------------------------------

### Summarize API Endpoint (POST)

Source: https://summarize.sh/docs/chrome-extension

The API endpoint on the daemon used to initiate a summarization request. It accepts a payload containing the URL, title, and extracted text.

```javascript
POST /v1/summarize
Authorization: Bearer <token>

{
  "url": "string",
  "title": "string",
  "text": "string"
}
```

--------------------------------

### POST /v1/summarize

Source: https://summarize.sh/docs/chrome-extension

Submits content to be summarized. Supports various customization options for the summarization process.

```APIDOC
## POST /v1/summarize

### Description
Submits content to be summarized. Supports various customization options for the summarization process.

### Method
POST

### Endpoint
/v1/summarize

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Parameters
#### Request Body
- **url** (string) - Required - The URL of the content to summarize.
- **title** (string | null) - Optional - A title for the content.
- **model** (string) - Optional - The summarization model to use (e.g., `auto`, `free`, `openai/gpt-5-mini`).
- **length** (string) - Optional - Desired length of the summary (e.g., `short`, `xl`, `20k`).
- **language** (string) - Optional - The language of the content (e.g., `auto`, `en`, `de`, `pt-BR`).
- **prompt** (string) - Optional - A custom instruction prefix for the summarization.
- **mode** (string) - Optional - The mode of operation (`auto`, `page`, `url`). Defaults to `"auto"`.
- **maxCharacters** (number | null) - Optional - Caps URL-mode extraction before summarization; ignored for extract-only unless explicitly provided.
- **format** (string) - Optional - Output format (`text` or `markdown`). Defaults to `"text"`.
- **markdownMode** (string) - Optional - Markdown processing mode (`readability`, `auto`, `llm`, `off`). Only when `format: "markdown"`.
- **preprocess** (string) - Optional - Markitdown/HTML preprocessing option (`off`, `auto`, `always`).
- **extractOnly** (boolean) - Optional - If true, returns extracted content without summarizing. Requires `mode: "url"`.
- **text** (string) - Required for `mode: "page"`; optional for `auto` - The text content to process.
- **truncated** (boolean) - Optional - Indicates if the extracted `text` was shortened.

### Request Example
```json
{
  "url": "https://example.com",
  "model": "auto",
  "length": "short",
  "language": "en"
}
```

### Response
#### Success Response (200)
- **ok** (boolean) - Indicates success.
- **id** (string) - The unique identifier for the summarization job.

#### Response Example
```json
{
  "ok": true,
  "id": "summary-12345"
}
```
```

--------------------------------

### Cache Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Configuration options for the on-disk SQLite cache in summarize_sh. This includes enabling/disabling the cache, setting size limits, time-to-live, and paths for both general content and media files. It also details the verification method for cached media.

```json
{
  "cache": {
    "enabled": true,
    "maxMb": 512,
    "ttlDays": 30,
    "path": "~/.summarize/cache.sqlite",
    "media": {
      "enabled": true,
      "maxMb": 2048,
      "ttlDays": 7,
      "path": "~/.summarize/cache/media",
      "verify": "size"
    }
  }
}

```

--------------------------------

### POST /v1/agent

Source: https://summarize.sh/docs/chrome-extension

Interacts with an AI agent for conversational tasks or content processing. Supports SSE streaming by default, with JSON output available via `Accept` header or query parameter.

```APIDOC
## POST /v1/agent

### Description
Interacts with an AI agent for conversational tasks or content processing. Supports SSE streaming by default, with JSON output available via `Accept: application/json` or `?format=json` query parameter.

### Method
POST

### Endpoint
/v1/agent

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Parameters
#### Request Body
- **url** (string) - Required - The URL of the content to process.
- **title** (string | null) - Optional - A title for the content.
- **pageContent** (string) - Required - The extracted text content from the page.
- **cacheContent** (string) - Optional - Content used for cache key generation; defaults to `pageContent`.
- **messages** (Array<Message>) - Required - An array of messages in pi-ai format for the conversation.
- **model** (string) - Optional - The AI model to use.
- **length** (string) - Optional - Desired length of the response (e.g., `short`, `xl`, `20k`).
- **language** (string) - Optional - The language of the content (e.g., `auto`, `en`, `de`).
- **tools** (string[]) - Optional - A list of available tools for the agent.
- **automationEnabled** (boolean) - Optional - Enables or disables automation features.

### Request Example
```json
{
  "url": "https://example.com/article",
  "pageContent": "Extracted content of the article...",
  "messages": [
    {"role": "user", "content": "Summarize this article."}
  ],
  "model": "gpt-4o"
}
```

### Response
#### SSE Stream Events
- **event: chunk** - Data containing text chunks (`data: { "text": "..." }`).
- **event: assistant** - Data containing assistant's response details (`data: { ...assistant }`).
- **event: done** - Event signaling the completion of the agent interaction (`data: {}`).
- **event: error** - Event containing error message if an issue occurred (`data: { "message": "Error details" }`).
```

--------------------------------

### Set Default Output Language for Summaries

Source: https://summarize.sh/docs/config

Configure the default language for generated summaries. The 'auto' setting matches the source language, while specific language codes or names can enforce a particular output language.

```json
{
  "output": { "language": "auto" }
}
```

--------------------------------

### Environment Variable Configuration for Summarize Sh

Source: https://summarize.sh/docs/config

Configuration for setting environment variables within the summarize_sh project's JSON config file. This allows pre-setting API keys and other environment-specific variables, though process environment variables take precedence.

```json
{
  "env": {
    "OPENAI_API_KEY": "sk-...",
    "OPENROUTER_API_KEY": "sk-or-...",
    "FIRECRAWL_API_KEY": "...",
    "CUSTOM_FLAG": "1"
  }
}

```

```json
{
  "apiKeys": {
    "openai": "sk-...",
    "anthropic": "sk-ant-...",
    "google": "...",
    "openrouter": "sk-or-...",
    "xai": "...",
    "zai": "...",
    "apify": "...",
    "firecrawl": "...",
    "fal": "..."
  }
}

```

--------------------------------

### POST /v1/summarize

Source: https://summarize.sh/docs/website

Summarizes content from a given website URL. It fetches HTML, extracts article-like content, and can optionally convert it to Markdown or plain text. Supports various flags for controlling the extraction and summarization process.

```APIDOC
## POST /v1/summarize

### Description
This endpoint fetches the HTML of a given URL, extracts the main content, and normalizes it into clean text. It can also convert the content to Markdown using different modes or fall back to Firecrawl for extraction. It handles video content and podcast audio transcription.

### Method
POST

### Endpoint
`/v1/summarize`

### Parameters
#### Query Parameters
- **url** (string) - Required - The URL of the website to summarize.
- **format** (string) - Optional - The output format. Options: `text` (default), `md`.
- **markdownMode** (string) - Optional - The mode for Markdown conversion when `format` is `md`. Options: `readability` (default), `auto`, `llm`, `off`.
- **firecrawl** (string) - Optional - Controls Firecrawl usage. Options: `off`, `auto`, `always`.
- **preprocess** (string) - Optional - Controls markitdown usage. Options: `off`, `auto` (default), `always`.
- **videoMode** (string) - Optional - Mode for handling video inputs. Options: `auto` (default), `transcript`, `understand`.
- **language** (string) - Optional - The language to use for summarization. Defaults to `auto`.
- **forceSummary** (boolean) - Optional - Forces the LLM to run even if extracted content is short.
- **extractOnly** (boolean) - Optional - Extracts content but does not perform summarization.
- **json** (boolean) - Optional - Emits a single JSON object for the output.
- **verbose** (boolean) - Optional - Enables verbose logging for progress and extractor information.
- **metrics** (string) - Optional - Controls metrics output. Options: `on` (default), `off`, `detailed`.
- **timeout** (string) - Optional - Sets a timeout for the request (e.g., `30s`, `2m`, `5000ms`). Defaults to `2m`.

#### Request Body
*Note: The primary input is the `url` query parameter. Request body is typically empty for this endpoint unless specific configurations are passed.* 

### Request Example
```json
{
  "url": "https://example.com/article",
  "format": "md",
  "markdownMode": "llm",
  "firecrawl": "always"
}
```

### Response
#### Success Response (200)
- **content** (string) - The summarized content in the specified format.

#### Response Example
```json
{
  "content": "This is a summarized version of the article content..."
}
```

### Error Handling
- **400 Bad Request**: Invalid input parameters or URL.
- **500 Internal Server Error**: Server encountered an issue processing the request.
```

--------------------------------

### Restart Summarize Daemon Service

Source: https://summarize.sh/docs/chrome-extension

Restarts the Summarize daemon service. This can resolve issues where the daemon may have stopped responding or encountered an error.

```bash
summarize daemon restart
```

--------------------------------

### POST /v1/agent/history

Source: https://summarize.sh/docs/chrome-extension

Submits content and conversation history to the agent for processing, returning a list of messages. Primarily used for replaying or analyzing past interactions.

```APIDOC
## POST /v1/agent/history

### Description
Submits content and conversation history to the agent for processing, returning a list of messages. Primarily used for replaying or analyzing past interactions.

### Method
POST

### Endpoint
/v1/agent/history

### Authentication
Requires authentication via `Authorization: Bearer <token>` header.

### Parameters
#### Request Body
- **url** (string) - Required - The URL of the content associated with the history.
- **pageContent** (string) - Required - The extracted text content from the page.
- **cacheContent** (string) - Optional - Content used for cache key generation; defaults to `pageContent`.
- **model** (string) - Optional - The AI model to use.
- **length** (string) - Optional - Desired length of the response (e.g., `short`, `xl`, `20k`).
- **language** (string) - Optional - The language of the content (e.g., `auto`, `en`, `de`).
- **automationEnabled** (boolean) - Optional - Enables or disables automation features.

### Request Example
```json
{
  "url": "https://example.com/chat",
  "pageContent": "Chat transcript content...",
  "model": "gpt-3.5-turbo"
}
```

### Response
#### Success Response (200)
- **ok** (boolean) - Indicates success.
- **messages** (Array<Message>) - A list of messages representing the conversation history or agent's processed output.

#### Response Example
```json
{
  "ok": true,
  "messages": [
    {"role": "user", "content": "Hello!"},
    {"role": "assistant", "content": "Hi there! How can I help you today?"}
  ]
}
```
```

--------------------------------

### Run End-to-End Tests Headless

Source: https://summarize.sh/docs/chrome-extension

Executes the end-to-end tests for the Chrome extension in headless mode. Note that headful mode is often more reliable for testing browser extensions.

```bash
HEADLESS=1 pnpm -C apps/chrome-extension test:e2e
```

--------------------------------

### Summarize SSE Events Endpoint

Source: https://summarize.sh/docs/chrome-extension

Endpoint to stream summarization results via Server-Sent Events (SSE).

```APIDOC
## GET /v1/summarize/{id}/events

### Description
Streams the summarization results for a given task ID using Server-Sent Events (SSE).

### Method
GET

### Endpoint
`http://127.0.0.1:8787/v1/summarize/<id>/events`

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier for the summarization task obtained from the POST /v1/summarize endpoint.

#### Query Parameters
None

#### Request Body
None

### Request Example
```bash
curl http://127.0.0.1:8787/v1/summarize/<summary_id>/events
```

### Response
#### Success Response (200)
- **Event Stream** - A stream of Server-Sent Events containing Markdown-formatted summary chunks.

#### Response Example
```
...
  data: ## Summary

  data: This is the first part of the summary.

  data: This is the second part of the summary.
...
```
```

--------------------------------

### POST /v1/summarize

Source: https://summarize.sh/docs/extract-only

Summarizes website content. Supports extract-only mode, markdown conversion, and slide detection.

```APIDOC
## POST /v1/summarize

### Description
Summarizes website content. Supports extract-only mode, markdown conversion, and slide detection.

### Method
POST

### Endpoint
/v1/summarize

### Parameters
#### Query Parameters
- **url** (string) - Required - The URL of the website to summarize.
- **extractOnly** (boolean) - Optional - If true, only extracts content without summarization. Defaults to false.
- **format** (string) - Optional - Output format. Supported values: "markdown".
- **markdownMode** (string) - Optional - Specifies the markdown conversion mode. Supported values: "readability", "auto", "llm".
- **length** (integer) - Optional - Guidance for summarization length (not used in extract-only mode).
- **maxExtractCharacters** (integer) - Optional - Maximum number of characters to extract.
- **slides** (boolean) - Optional - If true, enables slide detection for videos. Defaults to false.
- **firecrawl** (string) - Optional - Controls Firecrawl usage. Supported values: "always".

### Request Body
(Not applicable for this endpoint, parameters are sent via query string)

### Request Example
```json
{
  "url": "https://example.com",
  "extractOnly": true,
  "format": "markdown",
  "markdownMode": "readability"
}
```

### Response
#### Success Response (200)
- **summary** (string) - The summarized content or extracted text.
- **transcriptSegments** (array) - Timed transcript segments (if available).
- **transcriptTimedText** (array) - Timed text transcript (if available).
- **slidesData** (object) - Slide detection metadata (if slides=true).

#### Response Example
```json
{
  "summary": "This is a summary of the website content...",
  "transcriptSegments": [
    {
      "text": "Hello world",
      "start": 0.5,
      "end": 2.0
    }
  ],
  "slidesData": {
    "slides": [
      {
        "timestamp": 10.5,
        "imageUrl": "/path/to/slide.jpg"
      }
    ]
  }
}
```
```

--------------------------------

### Configuring yt-dlp Cookies for YouTube Summarization

Source: https://summarize.sh/docs/youtube

Provides instructions on how to pass browser cookies to `yt-dlp` when encountering a 403 error from YouTube. This is done by setting the `SUMMARIZE_YT_DLP_COOKIES_FROM_BROWSER` environment variable.

```bash
export SUMMARIZE_YT_DLP_COOKIES_FROM_BROWSER=chrome
```

```bash
export SUMMARIZE_YT_DLP_COOKIES_FROM_BROWSER="chrome:Profile 1"
```

--------------------------------

### Set Default Model Preset in Summarize SH

Source: https://summarize.sh/docs/config

Specify a default model preset to be used for all summarization tasks. This simplifies usage by avoiding the need to specify the model on the command line for common tasks.

```json
{
  "model": "fast"
}
```

--------------------------------

### Check Summarize Daemon Status

Source: https://summarize.sh/docs/chrome-extension

Checks the current status of the Summarize daemon service. This is useful for verifying if the daemon is running correctly.

```bash
summarize daemon status
```

--------------------------------

### Extracting Slide Screenshots from YouTube Videos

Source: https://summarize.sh/docs/youtube

Enables the extraction of slide screenshots from YouTube videos. Requires `ffmpeg` and `yt-dlp`. Scene detection is automatically tuned. The `--slides-ocr` flag enables Optical Character Recognition on the slides.

```bash
summarize "https://www.youtube.com/watch?v=..." --slides
```

```bash
summarize "https://www.youtube.com/watch?v=..." --slides --slides-ocr
```

--------------------------------

### Summarize API Endpoint

Source: https://summarize.sh/docs/chrome-extension

The Summarize API endpoint for processing web content.

```APIDOC
## POST /v1/summarize

### Description
Submits content to the Summarize daemon for processing and returns a summary ID.

### Method
POST

### Endpoint
`http://127.0.0.1:8787/v1/summarize`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **url** (string) - Required - The URL of the web page to summarize.
- **title** (string) - Required - The title of the web page.
- **text** (string) - Required - The extracted readable text content from the web page.

### Request Example
```json
{
  "url": "https://example.com",
  "title": "Example Domain",
  "text": "This is the content of the example domain..."
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the summarization task.

#### Response Example
```json
{
  "id": "<summary_id>"
}
```
```

--------------------------------

### YouTube Extraction Modes for Summarization

Source: https://summarize.sh/docs/youtube

Specifies different modes for extracting YouTube video content. 'auto' attempts multiple methods, 'web' focuses on web-based captions, 'no-auto' prioritizes creator captions, 'apify' uses the Apify service, and 'yt-dlp' leverages yt-dlp for downloading and transcription.

```bash
pnpm summarize -- --extract "https://www.youtube.com/watch?v=I845O57ZSy4&t=11s"
```

```bash
summarize -- --youtube auto
summarize -- --youtube web
summarize -- --youtube no-auto
summarize -- --youtube apify
summarize -- --youtube yt-dlp
```

--------------------------------

### Summarize Health Check

Source: https://summarize.sh/docs/chrome-extension

Endpoint to check the health status of the Summarize daemon.

```APIDOC
## GET /health

### Description
Checks if the Summarize daemon is running and responsive.

### Method
GET

### Endpoint
`http://127.0.0.1:8787/health`

### Parameters
None

### Request Example
```bash
curl http://127.0.0.1:8787/health
```

### Response
#### Success Response (200)
Indicates the daemon is healthy.

#### Response Example
```
OK
```
```
Tokens:
Raw