Bitterbot is a local-first personal AI agent for people who want an assistant that remembers ongoing projects and preferences across conversations. It runs on your own device or server, with a browser dashboard for chatting, reviewing memory activity and managing skills. It's open source under the MIT license.
Its main distinction is how it handles memory. Frequently used facts gain weight, less useful information fades, and a knowledge graph tracks connections between people and projects. Durable facts stay in the agent's context, while contradictory information updates earlier beliefs without discarding their history.
The "dream engine" reviews accumulated experience between conversations. It merges overlapping memories, extracts reusable skills from successful tasks, researches gaps and revisits relationships as context changes. It evaluates those cycles to guide later memory work. User-defined safety rules constrain changes to the agent's personality.
Bitterbot can browse the web, execute code and communicate through WhatsApp. Its peer-to-peer network lets agents exchange skills through a marketplace using USDC. Group and channel sessions can run in separate Docker sandboxes, and unknown message senders need pairing approval.
Local hosting doesn't mean fully offline operation. Model connections use API credentials for services such as Anthropic, OpenAI and NEAR AI Cloud; web research and peer exchange also involve network access. Windows runs through WSL2. Default outbound connections have documented controls for disabling them.
Claim this page and we'll verify you by hand. Bitterbot gets the verified badge, and you can upgrade the listing to be featured on localhosted. Proud to be listed? Put our badge on your site.
Want more people to find Bitterbot?Promote it
Something wrong or outdated on this page?
203Updated 4 days ago
Linux · Docker · Web#Code execution#Guardrails#Human approval
Frona is a self-hosted AI assistant for people who want agents to carry out tasks with controlled access to their servers, files, and accounts. Its agents can browse websites, run code, build applications, and delegate work to other agents. You interact through chat, while sandboxes restrict what each agent can access.
6.7KUpdated 1 month agoApache-2.0
macOS · Windows · Linux · Web#Agent Skills#Code execution#Multi-agent workflows
HanaAgent is an open source personal AI agent for everyday computer work, with a graphical interface, persistent memory and customizable personalities. It runs on macOS (Apple Silicon and Intel), Windows and Linux, and suits people who want help with office tasks as well as coding.
2.9KUpdated 1 week agoMIT
macOS · Windows · Linux · Docker · Web#Agent Skills#Code execution#GGUF
18.2KUpdated 5 months agoApache-2.0
macOS · Windows · Linux · Docker · Web#Guardrails#Human approval#Knowledge graphs
886Updated 2 weeks agoMIT
macOS · Windows · Linux · Docker · Web#Agent Skills#Code execution#Guardrails
233Updated 3 weeks ago
macOS · Linux · Docker · Web#Context compression#Human approval#Knowledge graphs
Moltis is a self-hosted personal AI agent server for people who want an assistant that remembers past conversations and can act on their behalf. It runs on your own hardware as a single Rust binary, with a browser interface and access through messaging apps. It's open source under the MIT license.
OpenFang runs scheduled AI agents on your own machine, with built-in task packages called Hands that report results to a dashboard. It's for people who want recurring research, monitoring or business tasks to continue without prompting an agent each time. The open source platform runs on macOS, Windows and Linux under the Apache 2.0 license.
PocketPaw is a self-hosted AI agent for people who want a personal assistant that can work with files, write code and use a browser on their own computer or server. It runs on macOS, Windows and Linux, with a web dashboard and access through messaging apps. It is open source under the MIT license.
Alfred Black is a self-hosted AI agent for solo business owners and consultants who need help with inbox triage, follow-ups, and keeping track of ongoing work. It runs continuously on a Linux server through Docker, with a web dashboard and conversations through Telegram, Slack, email, or web chat.