Skip to main content
Polly provides an AI assistant embedded in your LangSmith workspace to help you analyze traces, threads, prompts, and experiment results. This page explains how to enable Polly on a self-hosted LangSmith instance.
Self-hosted LangSmith is an add-on to the Enterprise plan. For more details, refer to Pricing. Contact our sales team if you want to get a license key to trial LangSmith in your environment.

​
Prerequisites

Before enabling Polly, complete the following setup steps:
  1. Install the base LangSmith platform:
  2. Enable LangSmith Deployment (agent deployment capabilities).

​
Components

Polly consists of two main components:
  • agentBootstrap: Job that deploys the LangSmith Deployment (agent) needed for Polly.
  • polly (agent): The main agent that powers Polly across LangSmith (observability, threads, prompt playground, experiments).

​
Enable Polly

Add the following to your values.yaml: 
backend:
  agentBootstrap:
    enabled: true

config:
  polly:
    enabled: true
    encryptionKey: "<your-encryption-key>"

​
Generate an encryption key

Polly requires a Fernet encryption key to securely store secrets. Generate one using Python: 
python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"
You can store the encryption key in a predefined Kubernetes secret using the polly_encryption_key parameter. See Use an existing secret for details.
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.