Pizza releases open-source event-driven AI agent
Start here
Pizza released an open-source AI agent, available as a desktop app and a command-line interface, that advances each turn through an event-driven state machine instead of a while-true loop. An immutable SQLite event log serves as the source for its interface, model context, and history tree, and the agent keeps conversation context across tasks. Pizza runs project agents in separate workspaces and can deploy to remote machines over SSH. Slack, Discord, Telegram, Lark, voice, and webhooks reach the same agent through a gateway, and apps connect via MCP. It supports 40+ model providers, including Anthropic, OpenAI, Google, DeepSeek, GLM, and Kimi. Installers are available for macOS Apple Silicon, Windows x64, and Linux x64, or it can be installed with npm. Channels, voice, and remote access require their corresponding service and account setup, and model usage and connected services may carry their own costs.
Generated from reports · updated 16 h ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AIShow HN: Pizza launches an event-driven AI agent built without a while-true loop
Pizza is an open-source AI agent with a desktop app and CLI that advances each turn through an event-driven state machine instead of a while-true loop, with an immutable SQLite event log as the source for its interface, model context, and history tree. The agent keeps conversation context across tasks, runs project agents in separate workspaces, and can deploy to remote machines over SSH, while Slack, Discord, Telegram, Lark, voice, and webhooks reach the same agent through a gateway. It supports 40+ model providers including Anthropic, OpenAI, Google, DeepSeek, GLM, and Kimi, connects apps via MCP, and ships installers for macOS Apple Silicon, Windows x64, and Linux x64, or can be installed with npm install -g @tomsun28/pizza.
Heat for this event
- Current, comparable range
- 6
- Peak, comparable range
- 10November 16 01:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.