Practical overview
When it fits
Consider General Bots when you want to configure document-aware bots and LLM workflows within a self-hosted platform that also includes business applications. Its documented local startup provisions a multi-service stack, so the setup is suited to users prepared to manage a substantial Rust-based project.
Who it is for
- Organizations seeking to run an AI agent platform on infrastructure they control.
- Developers building document-grounded assistants or LLM workflows that use tools and dialog scripts.
- Teams looking to combine agent workflows with business applications such as CRM, mail and calendar.
Documented uses
- Build a bot that indexes organizational documents and retrieves information from them.
- Expose a tool to an LLM and use dialog scripts to guide interactions.
- Run multiple AI agents and LLM workflows within a self-hosted platform.
- Keep bot scripts, configuration, knowledge-base documents and generated files in a structured bot directory.
Requirements & limits
The project describes itself as open source under the MIT license and provides a local self-hosting startup path. Its documented requirements are a recent stable Rust toolchain using edition 2021 and Git; mold is optional.
- The project is built in Rust, and its README requires a recent stable Rust toolchain for the documented local startup.
- The startup script builds botserver and botui and provisions PostgreSQL, Vault, MinIO, Valkey, Zitadel, Qdrant and a local llama.cpp server.
- Bots use a directory-based structure with BASIC dialog scripts, knowledge-base documents, configuration and generated-file storage.
- The README points builders to the BotBook for the full keyword reference and examples.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Prepare a local General Bots instance
Build and start the documented local platform stack.
Before you start
- A recent stable Rust toolchain using edition 2021
- Git
- mold is optional
Steps
Clone the General Bots repository with Git.
Change into the repository directory.
Run ./restart.sh to build botserver and botui and start them.
Expected outcome
The README lists local addresses for the Suite, Cloud, Login and API after startup. It says first boot downloads the stack and models.
Guide limitations
- The instructions describe local startup; they do not establish deployment requirements for other environments.
- First startup may take a few minutes while the stack and models download.
Structure a bot with documents and dialog scripts
Set up the documented bot directory components and connect a knowledge base, tool and dialog.
Before you start
- A General Bots instance with Drive
- A bot directory using the documented .gbai structure
- Documents to place in the bot knowledge-base directory
Steps
Create a bot directory with .gbai and add the documented .gbdialog, .gbkb, .gbot and .gbdrive components.
Use USE KB to index and retrieve from a knowledge base.
Use USE TOOL to expose a tool to the LLM.
Use TALK to reply to a user or CALL to run another dialog.
Expected outcome
The README describes a bot directory that the platform picks up automatically, with documents, scripts, configuration and generated-file storage.
Guide limitations
- The README points to the BotBook for the full keyword reference and worked examples.
- The structure and keywords do not specify the behavior of a particular custom workflow.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 4 October 2026
- Original website title
- generalbots/generalbots · Official homepage title · retained in its source language