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

CategoryIntegrations
AI FrameworksLangChain, LlamaIndex, CrewAI, AutoGen, Haystack, DSPy, Semantic Kernel, Vercel AI SDK
Visual BuildersLangflow, Flowise, Dify
Automationn8n, Zapier, Make.com, Activepieces, Temporal
Web ScrapingScrapy, Crawl4AI, Firecrawl, ScrapeGraphAI
DatabasesSupabase, Prisma, PlanetScale, Airtable
Developer ToolsVS Code, Cursor, Raycast, GitHub Copilot, Cloudflare Workers
Self-Hosted LLMsOpen WebUI, OpenAI GPT Actions
Every integration is open source, Apache 2.0 licensed, and maintained in the plasmate-labs organization.

Why This Matters

1. Drop-in Replacements

Many of our integrations are drop-in replacements for existing tools:

You don't need to rewrite your code. Just swap the import.

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:

  1. Add the Plasmate action to your Zap
  2. Input a URL
  3. Get clean, structured JSON
No API keys required — Plasmate runs locally.

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:

If there's a framework or tool you'd like to see integrated, open an issue or submit a PR.

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.