Installation

Requirements

• Python 3.10 or higher
• A supported model provider (AWS credentials for Bedrock, or an API key for others)

Install the SDK

pip install --extra-index-url https://elsai-agents.elsai.ai/root/ elsai-agents==0.3.5
pip install --extra-index-url https://core-packages.elsai.ai/root/elsai-model/ "elsai-model[bedrock]==2.1.1"

Agent() defaults to Amazon Bedrock — configure AWS credentials after install (see below). For other providers, swap the extra (for example "elsai-model[openai]==2.1.1") — see Model Providers.

Prebuilt tools package

Install ready-made agent tools (file I/O, shell, web search, AWS, multi-agent, and more):

pip install --extra-index-url https://elsai-agents.elsai.ai/root/ elsai-agents-tools==0.3.0

Some tools require optional extras (browser automation, Mem0 memory, diagram generation, etc.). See Prebuilt Tools for the full list and install commands.

Optional extras

Install additional system protocols or telemetry extensions:

# OpenTelemetry exporter (OTLP)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[otel]==0.3.2"

# Agent-to-Agent protocol (A2A) — requires 0.3.0+ for A2A Protocol v1.0
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[a2a]==0.3.5"

# Everything (including A2A and OTEL)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[all]==0.3.2"

Memory integrations

elsai memory features are optional extras on elsai-agents. Install only what your pipeline needs — see Memory Overview for details.

# Chat history shaping (trim, LRU, TTL, summarize, JSON persistence)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[elsai-chat-history]==0.3.5"

# Embedding models (Azure OpenAI, AWS Bedrock)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[elsai-embeddings]==0.3.5"

# Vector stores (Chroma, Pinecone, Weaviate)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[elsai-vectordb]==0.3.5"

# Full vector memory stack (chat-history + embeddings + vectordb)
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ "elsai-agents[elsai-memory]==0.3.5"

Extra	Powers
elsai-chat-history	Trim, LRU, TTL, summarization steps; persistence="elsai_json"
elsai-embeddings	Azure / Bedrock embedding clients for vector memory
elsai-vectordb	Chroma, Pinecone, Weaviate clients
elsai-memory	Similarity retrieval and semantic memory (all three above)

Native pipeline steps (ElsaiSlidingStep, ElsaiSummarizingStep) work with the base elsai-agents install and do not require these extras.

Virtual environment (recommended)

python -m venv .venv
source .venv/bin/activate      # Windows: .venv\Scripts\activate
pip install --extra-index-url https://elsai-agents.elsai.ai/root/ elsai-agents==0.3.5
pip install --extra-index-url https://core-packages.elsai.ai/root/elsai-model/ "elsai-model[bedrock]==2.1.1"

Verify installation

import elsai
print(elsai.__version__)

Configure your model provider

By default the SDK uses Amazon Bedrock with model ID global.anthropic.claude-sonnet-4-6 in us-west-2. Make sure you have AWS credentials configured:

aws configure
# or set environment variables
export AWS_ACCESS_KEY_ID=...
export AWS_SECRET_ACCESS_KEY=...
export AWS_DEFAULT_REGION=us-west-2
# AWS_REGION is also respected by BedrockModel

For other providers, see the Model Providers section.

Editor setup

Download Cursor, Copilot, or Claude Code configs from Editor Setup.