Tiger Agents For Work

Tiger Agents for Work

Tiger Agent is a production-ready library and CLI for building Slack-native agents for serious work. Tiger Agents can handle concurrent conversations with enterprise-grade reliability and (optionally) zero-code setup.

• Want to develop with Tiger Agent as quickly as possible? Jump to the Developer Quick Start.
• Want to know what makes Tiger Agent special? Continue to the Features.
• Want to dig into the details? Consume the docs.

Features

• Durable & Atomic Event Handling: PostgreSQL-backed event claiming ensures exactly-once processing with no duplicates, even under high concurrency and failure conditions
• Bounded Concurrency: Fixed worker pools prevent resource exhaustion while maintaining predictable performance under load
• Immediate Event Processing: Events are processed within milliseconds of arrival rather than waiting for polling cycles, providing real-time responsiveness
• Resilient Retry Logic: Automatic retry with visibility thresholds and cleanup of stuck or expired events
• Horizontal Scalability: Multiple instances can run simultaneously with coordinated work distribution across all instances
• AI-Powered Responses: Pydantic-AI Agents using the model of your choice with (optional) MCP server integrations
• Extensible Architecture: No coding required for basic bots and easy customization via Python for more specialized use cases
• Complete Observability: Full Logfire instrumentation provides detailed tracing of event flow, worker activity, and database operations

How does Tiger Agent provide all of these features? Read the [docs](/docs/README.md).

Developer Quick Start

0. Prerequisites

1. uv
2. docker
3. An ANTHROPIC_API_KEY

1. Clone the repo and install the dependencies.

# clone the repo
git clone https://github.com/timescale/t