60+ Integrations: The Plasmate Ecosystem
What started as a Rust CLI for fetching web pages has grown into something much bigger. Today, Plasmate has 60+ integration repos spanning every major AI framework, automation platform, and developer tool.
This isn't just about numbers. It's about being able to drop Plasmate into whatever stack you're already using and immediately get 10-100x token compression on web content.
The Ecosystem at a Glance
| Category | Integrations |
|---|---|
| AI Frameworks | LangChain, LlamaIndex, CrewAI, AutoGen, Haystack, DSPy, Semantic Kernel, Vercel AI SDK |
| Visual Builders | Langflow, Flowise, Dify |
| Automation | n8n, Zapier, Make.com, Activepieces, Temporal |
| Web Scraping | Scrapy, Crawl4AI, Firecrawl, ScrapeGraphAI |
| Databases | Supabase, Prisma, PlanetScale, Airtable |
| Developer Tools | VS Code, Cursor, Raycast, GitHub Copilot, Cloudflare Workers |
| Self-Hosted LLMs | Open WebUI, OpenAI GPT Actions |
Why This Matters
1. Drop-in Replacements
Many of our integrations are drop-in replacements for existing tools:
- crawl4ai-plasmate — Same API as Crawl4AI, but with SOM output
- firecrawl-plasmate — Firecrawl-compatible interface, runs locally
- scrapy-plasmate — Middleware that replaces Splash/Playwright
2. Token Compression Everywhere
Whether you're using LangChain, n8n, or a custom Temporal workflow, you get the same benefit: 10-100x fewer tokens for web content.
At $3 per million input tokens, that's the difference between a hobby project and production costs.
3. Framework Compatibility
Building with Langflow? There's a component. Using Flowise? There's a node. Want it in your VS Code sidebar? Extension is ready.
The goal is to meet developers where they are, not force them into a new workflow.
Highlighted Integrations
LangChain & LlamaIndex
The most popular AI frameworks now have native Plasmate loaders:
from langchain_plasmate import PlasmateSOMLoader
loader = PlasmateSOMLoader(url="https://docs.stripe.com/api")
docs = loader.load()
# 54x fewer tokens than raw HTML
Zapier & Make.com
7,000+ apps are now one step away from structured web content:
- Add the Plasmate action to your Zap
- Input a URL
- Get clean, structured JSON
Supabase & Prisma
Store web content with vector search:
from supabase_plasmate import PlasmateSupabase
client = PlasmateSupabase(supabase_url, supabase_key)
await client.fetch_and_store("https://example.com", "web_content")
results = await client.semantic_search("pricing details", "web_content")
GitHub Copilot SDK
The just-released Copilot SDK now has Plasmate tools:
import { registerPlasmateSkill } from 'copilot-plasmate'
// Copilot can now fetch and understand any web page
registerPlasmateSkill(copilot)
What's Next
We're continuing to expand coverage:
- Graphiti — Real-time knowledge graphs from web content
- Temporal — Enterprise workflow orchestration
- Open WebUI — Self-hosted LLM interfaces
Get Started
The full list of integrations is maintained at awesome-plasmate.
To install Plasmate:
curl -fsSL https://plasmate.app/install.sh | sh
Or via package managers:
cargo install plasmate
pip install plasmate
npm install plasmate
The web wasn't built for AI agents. SOM was. And now it's available everywhere you build.