Skip to main content
The LangSmith Agent Builder Slack app integrates your agents with Slack for secure, context-aware communication inside your Slack workspace. After installation, your agents will be able to:
  • Send direct messages.
  • Post to channels.
  • Read thread messages.
  • Reply in threads.
  • Read conversation history.

​
How to install

To install the LangSmith Agent Builder for Slack:
  1. Navigate to Agent Builder in your LangSmith workspace.
  2. Create or edit an agent.
  3. Add Slack as a trigger or enable Slack tools.
  4. When prompted, authorize the Slack connection.
  5. Follow the OAuth flow to grant permissions to your Slack workspace.
The app will be installed automatically when you complete the authorization.

​
Permissions

The LangSmith Agent Builder requires the following permissions to your Slack workspace:
  • Send messages - Send direct messages and post to channels
  • Read messages - Read channel history and thread messages
  • View channels - Access basic channel information
  • View users - Look up user information for messaging
These permissions enable agents to communicate effectively within your Slack workspace.

​
Privacy policy

The LangSmith Agent Builder Slack app collects, manages, and stores third-party data in accordance with our privacy policy. For full details on how your data is handled, please see our privacy policy.

​
AI components and disclaimers

The LangSmith Agent Builder uses large language models (LLMs) to power AI agents that interact with users in Slack. While these models are powerful, they have the potential to generate inaccurate responses, summaries, or other outputs.

​
What you should know

  • AI-generated content: All responses from agents are generated by AI and may contain errors or inaccuracies. Always verify important information.
  • Data usage: Slack data is not used to train LLMs. Your workspace data remains private and is only used to provide agent functionality.
  • Transparency: The Agent Builder is transparent about the actions it will take once added to your workspace, as outlined in the permissions section above.

​
Technical details

The Agent Builder uses the following approach to AI:
  • Model: Uses LLMs provided through the LangSmith platform
  • Data retention: User data is retained according to LangSmith’s data retention policies
  • Data tenancy: Data is handled according to your LangSmith organization settings
  • Data residency: Data residency follows your LangSmith configuration
For more information about AI safety and best practices, see the Agent Builder documentation.

​
Pricing

The LangSmith Agent Builder Slack app itself does not have any direct pricing. However, agent runs and traces are billed through the LangSmith platform according to your organization’s plan. For current pricing information, see the LangSmith pricing page.
Edit the source of this page on GitHub.
Connect these docs programmatically to Claude, VSCode, and more via MCP for real-time answers.