nanobot
Product status: Status not recorded
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
No status check date recorded
HKUDS self-hosted Python agent with a WebUI, scheduled tasks, memory and chat integrations; models and external channels are configured separately. No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
Product preview
Decision notes
Official-document research
Based on the cited documentation. We did not use the product, purchase a plan, or run quality, speed or reliability tests in this review.
Consider it for
- Operators comfortable maintaining agent credentials and permissions
Choose another approach if
- Assuming default settings isolate an untrusted agent or make all model use free
Pricing and conditions
Documented terms
MIT source use has no project license charge, but nanobot supplies no model account. The README explicitly separates model fees/local hardware and says persistent disks for the Render deployment require a paid service. Chat, web search, tracing, hosting and API providers can bill separately. No combined subscription price, model unit rate, currency, tax or regional total was verified.
Pricing source read:
Pricing or terms sourceCode, model and service licensing
MIT permits code use, modification and redistribution with copyright/permission notices and no warranty. It does not grant model-weight, extension, channel, input or generated-output rights. Local sessions/memory are operator-managed, but content routed to model/chat/web/tracing providers follows those providers' processing, retention and training terms. No single hosted no-training, deletion schedule or commercial output indemnity was established. The official wiki was linked from the repository but returned no readable body, so current-source documentation supplies this review's evidence.
Basis and limits of the recorded score
The existing 4.5 editorial score is retained without its original scoring worksheet. This document review did not re-score or measure performance.
Overview
HKUDS nanobot is a self-hosted Python agent framework used through a terminal, WebUI, API or connected chat channel. It supports saved conversations, memory, scheduled tasks and MCP tools, with inference supplied by a configured provider or local server. The README distinguishes current source from released documentation; verify the installed version before assuming a newly documented feature exists.
The operator manages the agent's credentials, stored sessions and tools. Configuration documents workspace restriction as off and shell execution as on by default; workspace checks alone are not an OS sandbox. Explicit process sandboxing on supported systems does not restrict network access. A selected model, chat platform, web tool or tracing service can receive data separately; self-hosting is not a universal privacy contract.
Sources
The factual claims in this review were read from the pages below. Each date is when an editor last opened the source, not when the product changed.
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
- Vendor documentation Checked 2026-10-15
Pros / Cons
Pros
- Self-managed sessions and tool configuration
- MIT code with model-provider choice
Cons
- Default tool settings require review
- Models, channels and hosting add costs and data paths
Who it's for
Personal scheduled workflows
Controlled chat-based tool access
Bottom line
Use as an operator-managed agent framework; review shell/workspace/network boundaries and each provider before exposing the WebUI or chat gateway.
Key Features
Tags
Compare nanobot with alternatives
Frequently asked questions
Is nanobot currently available?
No lifecycle review is recorded for this listing. Its inclusion in the directory does not confirm current availability; check the vendor before relying on it.
What is nanobot?
nanobot is a Agents & Developer Tooling tool developed by HKUDS. HKUDS self-hosted Python agent with a WebUI, scheduled tasks, memory and chat integrations; models and external channels are configured separately.
What costs and terms should I check for nanobot?
MIT source use has no project license charge, but nanobot supplies no model account. The README explicitly separates model fees/local hardware and says persistent disks for the Render deployment require a paid service. Chat, web search, tracing, hosting and API providers can bill separately. No combined subscription price, model unit rate, currency, tax or regional total was verified.
Is nanobot open source?
MIT permits code use, modification and redistribution with copyright/permission notices and no warranty. It does not grant model-weight, extension, channel, input or generated-output rights. Local sessions/memory are operator-managed, but content routed to model/chat/web/tracing providers follows those providers' processing, retention and training terms. No single hosted no-training, deletion schedule or commercial output indemnity was established. The official wiki was linked from the repository but returned no readable body, so current-source documentation supplies this review's evidence.
What are the best alternatives to nanobot?
Notable Agents & Developer Tooling alternatives include Firecrawl, Browser Use, Daytona. Browse all 33 tools in our Agents & Developer Tooling category for a full overview.
Was this listing useful?
Basic Info
- Pricing
- Free
MIT source use has no project license charge, but nanobot supplies no model account. The README explicitly separates model fees/local hardware and says persistent disks for the Render deployment require a paid service. Chat, web search, tracing, hosting and API providers can bill separately. No combined subscription price, model unit rate, currency, tax or regional total was verified.
- Editorial score
- ★ 4.5 / 5 How we score
- Platforms
- CLIWebSelf-hosted
- Last verified
- Not recorded
