n8n-nodes-anakin-org
Advanced tools
n8n node for Anakin API - web scraping, AI search, and intelligent data extraction
This is an n8n community node that lets you use the Anakin API in your n8n workflows.
Anakin provides powerful web scraping, AI-powered search, and intelligent data extraction capabilities. This node handles all the complexity of job submission and polling automatically.
n8n-nodes-anakin-org in the Package name fieldNavigate to your n8n installation folder and run:
npm install n8n-nodes-anakin-org
Then restart n8n.
Before using the Anakin node, you need to set up your API credentials:
https://api.anakin.io)The Anakin node supports 21 operations covering the entire Anakin API: web scraping and AI search, site discovery (Map/Crawl), Wire automation actions, website monitoring, AI visibility comparison, browser session management, and AI-driven browser tasks.
Extract content and structured data from any website.
Trigger → Anakin (Scrape URL) → Process Data
Configuration:
https://example.com/product-pageus (optional)false (optional)300 seconds (optional)3 seconds (optional)Output:
{
"success": true,
"operation": "scrapeUrl",
"request_id": "req_123456",
"url": "https://example.com/product-page",
"status": "completed",
"html": "...",
"markdown": "...",
"generatedJson": {
// Structured data extracted from the page
}
}
Perform AI-powered searches using Perplexity AI. Get instant answers with citations.
Trigger → Anakin (Search) → Process Results
Configuration:
What are the latest trends in AI?5 (optional, default: 5)Output:
{
"success": true,
"operation": "search",
"query": "What are the latest trends in AI?",
"answer": "Based on recent developments...",
"results": [
{
"title": "AI Trends 2026",
"url": "https://example.com/ai-trends",
"content": "Summary of the article...",
"score": 0.95
}
],
"count": 5
}
Use Cases:
Advanced multi-stage AI pipeline that automatically:
Trigger → Anakin (Agentic Search) → Process Structured Data
Configuration:
Find the pricing plans for top 5 CRM softwaretrue (optional, more reliable)600 seconds (optional)5 seconds (optional)Output:
{
"success": true,
"operation": "agenticSearch",
"job_id": "job_789",
"status": "completed",
"query": "Find the pricing plans for top 5 CRM software",
"perplexity_answer": "Here are the top CRM solutions...",
"citations": [
{"url": "https://salesforce.com/pricing", "title": "Salesforce Pricing", "source_index": 0}
],
"scraped_data": [
{
"source_url": "https://salesforce.com/pricing",
"source_index": 0,
"data": {
"plans": [...],
"features": [...]
}
}
],
"chatgpt_schema": {
"type": "object",
"properties": {...}
},
"chatgpt_structured_data": {
"crm_platforms": [...]
},
"chatgpt_summary": "Comprehensive analysis of CRM pricing..."
}
Use Cases:
Beyond Scrape URL, Search, and Agentic Search, the node exposes the rest of the Anakin API as Operations. Full field-by-field defaults are in Configuration Options by Operation below; endpoints are listed in API Endpoints Used.
credential_id usable with the two action operations above. Synchronous.failed per-source result is still returned as data, not thrown as an error.| Option | Description | Default |
|---|---|---|
| URL | The website URL to scrape | Required |
| Max Wait Time | Maximum seconds to wait for completion | 300 |
| Poll Interval | Seconds between status checks | 3 |
| Country Code | Proxy country code (e.g., us, uk, de) | us |
| Force Fresh | Bypass cache and force fresh scrape | false |
| Option | Description | Default |
|---|---|---|
| Search Query | The question or query to search | Required |
| Max Results | Maximum number of results to return | 5 |
| Option | Description | Default |
|---|---|---|
| Search Prompt | The search prompt for analysis | Required |
| Use Browser | Use browser for scraping (more reliable) | true |
| Max Wait Time | Maximum seconds to wait for completion | 600 |
| Poll Interval | Seconds between status checks | 5 |
| Data Schema (JSON) | Optional JSON schema for structured data extraction | None |
| Option | Description | Default |
|---|---|---|
| URL | The starting URL for link discovery | Required |
| Limit | Maximum number of URLs to return overall | 100 |
| Depth | How many link-hops from the starting URL to follow | 2 |
| Limit Per Level | Maximum URLs collected per depth level | 100 |
| Include Subdomains | Include URLs on subdomains of the starting host | false |
| Include External Links | Also collect (but not follow) external links | false |
| Use Browser | Render with a headless browser (for SPAs) | false |
| Search Filter | Optional keyword filter on path/title | None |
| Max Wait Time | Maximum seconds to wait for completion | 300 |
| Poll Interval | Seconds between status checks | 3 |
| Option | Description | Default |
|---|---|---|
| URL | The starting URL to crawl | Required |
| Max Pages | Hard cap on pages fetched | 10 |
| Depth | Link-hops from the starting URL to follow | 1 |
| Country Code | Proxy egress country code | us |
| Use Browser | Render each page in a headless browser (for SPAs) | false |
| Include Patterns | Comma-separated glob/regex patterns; only matching URLs are fetched | None |
| Exclude Patterns | Comma-separated glob/regex patterns; matching URLs are skipped | None |
| Session ID | Saved browser-session ID for login-protected sites | None |
| Session Name | Saved browser-session name | None |
| Max Wait Time | Maximum seconds to wait for completion | 300 |
| Poll Interval | Seconds between status checks | 3 |
| Option | Description | Default |
|---|---|---|
| Query | The intent in natural language | Required |
| Limit | Maximum number of candidate actions to return | 5 |
| Option | Description | Default |
|---|---|---|
| Catalog Slug | Site slug to inspect (e.g. "walmart"); leave empty to list all catalogs | None |
Both operations share the same fields — they hit the identical POST /wire/task endpoint and are split only for read/write safety labeling.
| Option | Description | Default |
|---|---|---|
| Action ID | The Wire action to run (from Wire: Discover Actions or Wire: Browse Catalog) | Required |
| Params (JSON) | The action's input parameters; shape depends on the action | None |
| Credential ID | Required when the action's auth_mode is "required" | None |
| Identity ID | Optional identity selector (alternative to Credential ID) | None |
| Max Wait Time | Maximum seconds to wait when the action runs asynchronously | 300 |
| Poll Interval | Default seconds between status checks (the API's own pacing hint is honored when provided) | 3 |
| Option | Description | Default |
|---|---|---|
| Catalog ID | Optional — restrict to identities for a single catalog | None |
| Option | Description | Default |
|---|---|---|
| Catalog Slug | The catalog to sign in to (e.g. "neb") | Required |
| Login Params (JSON) | Login fields defined by the catalog (e.g. email/password) | None |
| Identity Name | Optional name for the identity | None |
| Source ID | Optional 1Password identity-source ID (alternative to Login Params) | None |
| Source Ref (JSON) | Optional 1Password item locator { vault_id, item_id, fields } | None |
| Option | Description | Default |
|---|---|---|
| Website URL | The site to build an action for | Required |
| Goal | Natural-language description of what the action should do or extract | Required |
| Catalog ID | Optional — attach to an existing catalog instead of creating one | None |
| Visibility | Action visibility: Private or Public | private |
| Force | Build even if similar actions already exist for the domain | false |
| Option | Description | Default |
|---|---|---|
| URL | The URL to watch | Required |
| Interval (Minutes) | Check frequency in minutes (minimum 15) | 60 |
| Scope | Page, Site, or Wire | page |
| Watch Mode | Full Page or Specific Data (AI-extracted fields) | full_page |
| Watch Format | Format compared in full_page mode: Markdown, HTML, or Cleaned HTML | markdown |
| Output Schema (JSON) | JSON Schema of the fields to track; required when Watch Mode is Specific Data | None |
| AI Mode | Filter trivial noise and summarize real changes with AI (+1 credit/check) | false |
| AI Goal | Natural-language description of which changes count as meaningful | None |
| Use Browser | Render checks with a stealth headless browser | false |
| Country Code | Two-letter proxy country code | us |
| Session ID | Saved browser-session ID for login-protected pages | None |
| Is Active | Start running immediately | true |
| Expires At | Optional end date (ISO 8601 or YYYY-MM-DD) | None |
| Alert Webhook URL | Webhook URL that receives signed change alerts | None |
| Alert Emails | Comma-separated email recipients for change alerts | None |
| Max Pages | Site scope: max pages crawled per run | API default |
| Max Depth | Site scope: crawl depth 1-5 | API default (2) |
| Include Patterns | Site scope: comma-separated patterns/URLs to track | None |
| Exclude Patterns | Site scope: comma-separated patterns to skip | None |
| Wire Action ID | Wire scope (required there): the action run each check | None |
| Wire Catalog Slug | Wire scope: catalog slug of the Wire site | None |
| Wire Credential ID | Wire scope: credential ID when the action needs auth | None |
| Wire Params (JSON) | Wire scope: parameters passed to the action each check | None |
| Wire Watch Paths | Wire scope: comma-separated JSON paths to diff | None |
| Option | Description | Default |
|---|---|---|
| Monitor ID | Optional — fetch just this monitor instead of the full list | None |
| Option | Description | Default |
|---|---|---|
| Monitor ID | The monitor ID | Required |
| Option | Description | Default |
|---|---|---|
| Monitor ID | The monitor ID | Required |
| Action | Pause, Resume, Run Now, or Delete | pause |
| Option | Description | Default |
|---|---|---|
| Query | The question to ask every AI engine (max 2000 characters) | Required |
| Sources | Comma-separated engine slugs to query; leave empty for all enabled engines | None |
| Country Code | Two-letter ISO country for the search geography | us |
| Include Full Content | Include each engine's raw full answer in the results (large) | false |
| Max Wait Time | Maximum seconds to wait for completion | 180 |
| Poll Interval | Seconds between status checks | 3 |
No configurable fields.
| Option | Description | Default |
|---|---|---|
| Domain | Optional — filter to sessions for one website domain | None |
| Option | Description | Default |
|---|---|---|
| Session ID | The session ID to delete | Required |
| Option | Description | Default |
|---|---|---|
| Prompt | The task in natural language; never include passwords or secrets | Required |
| URL | Navigate here before starting | None |
| Session ID | Saved browser-session ID so the task runs logged in | None |
| Max Steps | Cap on agent steps | API default |
| Timeout (Ms) | Task timeout in milliseconds (server caps runs at ~330s regardless) | API default |
| Output Schema (JSON) | JSON Schema for the result | None |
| Max Wait Time | Maximum seconds to wait for completion | 360 |
| Poll Interval | Seconds between status checks | 3 |
Scrape URL, Agentic Search, Map, Crawl, Wire: Run Read/Write Action (when the action isn't synchronous), and AI Visibility Search follow the same pattern:
Search, Wire: Discover Actions, Wire: Browse Catalog, Wire: List Identities, Wire: Sign In, Wire: Request New Action, Create Monitor, List Monitors, Get Monitor Changes, Control Monitor, AI Visibility Sources, List Sessions, and Delete Session return immediately — no polling required.
The node handles all the complexity of:
The node will throw an error if:
You can enable Continue on Fail in the node settings to handle errors gracefully.
POST /v1/request - Submit scraping jobGET /v1/request/{id} - Check job statusPOST /v1/search - Perform AI search (synchronous)POST /v1/agentic-search - Submit agentic search jobGET /v1/agentic-search/{jobId} - Check agentic search statusPOST /v1/map - Submit mapping jobGET /v1/map/{jobId} - Check job statusPOST /v1/crawl - Submit crawl jobGET /v1/crawl/{jobId} - Check job statusGET /v1/wire/resolve - Discover candidate actions from a natural-language intentGET /v1/wire/catalog / GET /v1/wire/catalog/{slug} - Browse the Wire catalogPOST /v1/wire/task - Run a Wire action (used by both Run Read Action and Run Write Action)GET /v1/wire/jobs/{jobId} - Check an async Wire action's job statusGET /v1/wire/identities - List saved identities/credentialsPOST /v1/wire/login - Sign in to a credentials-mode sitePOST /v1/wire/build-request - Request a new action for an unsupported sitePOST /v1/monitors - Create a monitorGET /v1/monitors / GET /v1/monitors/{id} - List monitors / fetch oneGET /v1/monitors/{id}/changes - Get a monitor's detected changesPOST /v1/monitors/{id}/pause / /resume / /run - Control a monitorDELETE /v1/monitors/{id} - Delete a monitorGET /v1/ai-visibility/sources - List available AI answer enginesPOST /v1/ai-visibility/search - Submit a searchGET /v1/ai-visibility/search/{search_id} - Check search statusGET /v1/sessions - List saved sessionsDELETE /v1/sessions/{id} - Delete a saved sessionPOST /v1/ai/evaluate - Submit an AI browser task (async)GET /v1/ai/jobs/{workflow_id} - Check task status# Clone the repository
git clone https://github.com/Anakin-Inc/anakin-n8n.git
cd anakin-n8n
# Install dependencies
npm install
# Build the node
npm run build
# Link for local development
npm link
cd ~/.n8n/custom
npm link n8n-nodes-anakin-org
n8n-nodes-anakin-org/
├── credentials/
│ └── AnakinScraperApi.credentials.ts
├── nodes/
│ └── AnakinScraper/
│ └── AnakinScraper.node.ts
├── package.json
└── README.md
For issues, questions, or contributions:
MIT
schema parameter already supported by the underlying APIFunction() constructor with sleep from n8n-workflowjobId extraction bug in Scrape URL operationFAQs
n8n node for Anakin API - web scraping, AI search, and intelligent data extraction
We found that n8n-nodes-anakin-org demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 2 open source maintainers collaborating on the project.
Research
/Security News
The compromise affects MemTensor's MemOS, an open source memory framework for large language models (LLMs) and AI agents. Both npm package @memtensor/memos-cloud-openclaw-plugin and the PyPI package MemoryOS are compromised. They drop cross-platform Go binaries that exfiltrate developer secrets.
Security News
Lovable’s OJ rewrites Vite’s dev server in Rust, reducing memory use and preview times as AI lowers the cost of open source reimplementation.
Security News
It has been one year since Shai-Hulud made its first appearance on npm.