DOCS

Migration Guide

Migrate to Skimly v2.0+ for production-grade streaming, tool calling, and Anthropic-compatible interfaces. Switch from provider SDKs or upgrade from Skimly v1.x in minutes.

Why Migrate?

v2.0 Features: Streaming, tool calling, TypeScript types, Anthropic-compatible interfaces.
Provider flexibility: Switch between OpenAI and Anthropic with one line change.
Cost savings: Smart Truncation + blob storage reduces token usage by 35-85%.

Skimly v1.x → v2.0+

Already using Skimly v1.x? Upgrade to v2.0+ for streaming, enhanced TypeScript support, and Anthropic-compatible interfaces:

v1.x (Legacy)

// Node.js v1
import { fromEnv } from '@skimly/sdk'
const client = fromEnv()

await client.chat({
  provider: 'openai',
  model: 'gpt-4o-mini',
  messages: [{ role: 'user', content: 'Hello!' }]
})

# Python v1
from skimly import Skimly
client = Skimly.from_env()

client.chat({
  'provider': 'anthropic',
  'model': 'claude-3-sonnet',
  'messages': [{'role': 'user', 'content': 'Hello!'}]
})

v2.0+ (Current)

// Node.js v2.0+
import { SkimlyClient } from '@skimly/sdk'
const client = new SkimlyClient({ 
  apiKey: process.env.SKIMLY_KEY! 
})

await client.messages.create({
  provider: 'openai',
  model: 'gpt-4o-mini',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Hello!' }]
})

# Python v2.0+
from skimly import AsyncSkimlyClient

async with AsyncSkimlyClient.from_env() as client:
    await client.messages.create({
        'provider': 'anthropic',
        'model': 'claude-3-sonnet',
        'max_tokens': 1024,
        'messages': [{'role': 'user', 'content': 'Hello!'}]
    })

🚀 v2.0 New Features

  • • Streaming: client.messages.stream() with real-time SSE
  • • Anthropic Interface: messages.create() matching official SDK
  • • Smart Truncation: Up to 99% savings on verbose AI responses
  • • Full TypeScript: Comprehensive types and error handling
  • • Async Python: AsyncSkimlyClient with context managers

Provider SDKs → Skimly

Migration Overview

The migration involves three main changes:

1. API Key

Replace OPENAI_API_KEY or ANTHROPIC_API_KEY with SKIMLY_KEY

2. SDK Import

Replace provider SDKs with @skimly/sdk (Node) or skimly (Python)

3. API Calls

Replace .chat.completions.create() or .messages.create() with client.chat()

Node.js Migration

OpenAI → Skimly

Replace direct OpenAI SDK calls with Skimly's unified interface:

Before (OpenAI SDK)

import OpenAI from "openai"

const openai = new OpenAI({ 
  apiKey: process.env.OPENAI_API_KEY 
})

const resp = await openai.chat.completions.create({
  model: "gpt-4o-mini",
  messages: [{ 
    role: "user", 
    content: "Summarize this email" 
  }]
})

After (Skimly SDK)

import { SkimlyClient } from "@skimly/sdk"

const client = new SkimlyClient({ 
  apiKey: process.env.SKIMLY_KEY! 
})

const resp = await client.messages.create({
  provider: "openai",
  model: "gpt-4o-mini",
  max_tokens: 1024,
  messages: [{ 
    role: "user", 
    content: "Summarize this email" 
  }]
})

Anthropic → Skimly

Replace direct Anthropic SDK calls with Skimly's unified interface:

Before (Anthropic SDK)

import Anthropic from "@anthropic-ai/sdk"

const anthropic = new Anthropic({ 
  apiKey: process.env.ANTHROPIC_API_KEY 
})

const resp = await anthropic.messages.create({
  model: "claude-3-sonnet",
  messages: [{ 
    role: "user", 
    content: "Summarize this email" 
  }]
})

After (Skimly SDK)

import { SkimlyClient } from "@skimly/sdk"

const client = new SkimlyClient({ 
  apiKey: process.env.SKIMLY_KEY! 
})

const resp = await client.messages.create({
  provider: "anthropic",
  model: "claude-3-5-sonnet-20241022",
  max_tokens: 1024,
  messages: [{ 
    role: "user", 
    content: "Summarize this email" 
  }]
})

Python Migration

OpenAI → Skimly

Replace direct OpenAI SDK calls with Skimly's unified interface:

Before (OpenAI SDK)

from openai import OpenAI

client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))

resp = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Summarize this email"}]
)

After (Skimly SDK)

from skimly import AsyncSkimlyClient
import asyncio

async def main():
    async with AsyncSkimlyClient.from_env() as client:
        resp = await client.messages.create({
            'provider': 'openai',
            'model': 'gpt-4o-mini', 
            'max_tokens': 1024,
            'messages': [{'role': 'user', 'content': 'Summarize this email'}]
        })

asyncio.run(main())

Anthropic → Skimly

Replace direct Anthropic SDK calls with Skimly's unified interface:

Before (Anthropic SDK)

import anthropic

client = anthropic.Anthropic(
    api_key=os.getenv("ANTHROPIC_API_KEY")
)

resp = client.messages.create(
    model="claude-3-sonnet",
    messages=[{"role": "user", "content": "Summarize this email"}]
)

After (Skimly SDK)

from skimly import AsyncSkimlyClient
import asyncio

async def main():
    async with AsyncSkimlyClient.from_env() as client:
        resp = await client.messages.create({
            'provider': 'anthropic',
            'model': 'claude-3-5-sonnet-20241022',
            'max_tokens': 1024,
            'messages': [{'role': 'user', 'content': 'Summarize this email'}]
        })

asyncio.run(main())

What Changed

API Key

Replace OPENAI_API_KEY or ANTHROPIC_API_KEY with SKIMLY_KEY

API Calls

Use client.messages.create() (Anthropic-style interface) for all providers

Provider Flexibility

Switch between OpenAI and Anthropic with one line change: provider: "openai" → provider: "anthropic"

Blob Storage

Upload large content once and reference it in chat requests to save 60-90% on tokens

Migration Checklist

Install Skimly SDK v2.0+

npm install @skimly/sdk@^2.0.0 (Node) or pip install skimly>= (Python)

Update Environment Variables

Replace OPENAI_API_KEY or ANTHROPIC_API_KEY with SKIMLY_KEY

Update SDK Imports

Replace provider SDK imports with Skimly SDK imports

Update API Calls

Use client.messages.create() (Anthropic-style interface) and add max_tokens

Add Provider Field

Add provider: "openai" or provider: "anthropic" to all chat calls

Test Migration

Verify responses are identical to your original provider calls

(Optional) Remove Old Dependencies

Uninstall openai and/or @anthropic-ai/sdk if no longer needed

Bonus: Blob Storage for Big Context

Once migrated, you can start using Skimly's blob storage to dramatically reduce token usage for large, rarely-changing content like policies, documentation, or conversation threads.

Upload Content Once

// Node.js
const blob = await client.createBlob(
  '...very long text content...', 
  'text/plain'
)
console.log(blob.blob_id) // -> "b_abc123..."

# Python
blob_id = client.create_blob(
    "...very long text content...", 
    "text/plain"
)
print(blob_id) # -> "b_abc123..."

Reference in Chat

// Node.js
const resp = await client.chat({
  provider: 'openai',
  model: 'gpt-4o-mini',
  messages: [{
    role: 'user',
    content: [
      { type: 'text', text: 'Summarize this:' },
      { type: 'pointer', blob_id: blob.blob_id }
    ]
  }]
})

# Python
resp = client.chat(
    provider="anthropic",
    model="claude-3-sonnet",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "Summarize this:"},
            {"type": "pointer", "blob_id": blob_id}
        ]
    }]
)
Ready to migrate?

Follow the checklist above step by step. Migration to v2.0+ typically takes 15-30 minutes and unlocks streaming, enhanced TypeScript support, and up to 99% token savings with Smart Truncation.

Next Steps

Quickstart Guide

Learn how to use Skimly's features like blob storage and provider switching.

View quickstart →

API Reference

Complete endpoint documentation with all available options and parameters.

View API docs →