- Fix TLS certificate verification, sandboxed shell execution and workspace permissions for queued inputs.
- Keep idle maintenance from blocking active work or incorrectly refusing an update. Improve retry, reconnection and recovery behavior, and preserve native provider state through compaction.
- Add a file-context graph with file previews, local Markdown links and recent context snapshots. Switching topics keeps each conversation's context separate.
- Share syntax highlighting across code, source previews and read-only diffs, with complete source copying and plain-text fallbacks.
- Reduce unnecessary memory copies in file reads, session history and provider requests. Improve navigation, context-usage display, history loading and skill-source identification.
- Retain legacy history and proxy migration throughout 0.4.x, including original workspace ownership for unmigrated conversations.
nanobot by HKUDS
A lightweight, self-hosted personal agent runtime in Python that runs in a web UI, a terminal or chat apps, using shell, files, web, MCP and sub-agents, with long-term memory and scheduled automations.
Use it in: CLI Web app Slack, Teams, Telegram API / SDK Self-hosted Works with: MCP Local models Own API key Multi-agent Headless / CI
On this page
Usage data: latest readings Nov 19 to Nov 21, 2026, last checked Nov 21, 06:32 UTC. Date of each number
What nanobot does
nanobot is an open-source personal AI agent from HKUDS, first released in February 2026. You use it through a command-line tool, a web app, messaging apps, an SDK and self-hosted on macOS, Windows and Linux. Its documentation covers MCP servers, subagents or parallel agents, a non-interactive mode for scripts and CI, your own API key and local models. It can run models from Anthropic, OpenAI, OpenRouter, Amazon Bedrock and any OpenAI-compatible endpoint. It is free and open source (MIT); you pay only for the models it calls.
Key facts
| Maker | HKUDS |
|---|---|
| Type | Open-source personal AI agent |
| First released | February 2026 |
| License | MIT |
| Interfaces | CLI, Web app, Messaging apps, SDK, Self-hosted |
| Platforms | macOS, Windows, Linux |
| Main language | Python |
| Open issues and PRs | 872 |
Pricing
Free MIT-licensed self-hosted software; users supply their own model providers or local models.
Source: official pricing page, checked Oct 6, 2026. Prices change often; confirm before you buy.
Quick answers
- Free to start?Yes
- Open source?Yes (MIT)
- Runs onmacOS, Windows, Linux
- Where you use itCLI, Web app, Messaging apps, SDK, Self-hosted
- Supports MCP?Yes
- Model providersanthropic, openai, openrouter, amazon, any OpenAI-compatible, local
- Subagents / multi-agent?Yes
- Runs in scripts / CI?Yes
- Feature comparisonAll agents
- Closest alternativeOpenClaw · all alternatives
What's new in nanobot: November 2026 so far
November brought significant stability improvements, better context management, and enhanced file handling to nanobot. The release focused on fixing critical security and reliability issues while adding visual context tools.
- Fixed TLS certificate verification, sandboxed shell execution, and workspace permission handling for queued inputs.
- Added a file-context graph with file previews, local Markdown links, and recent context snapshots to track conversation context across topic switches.
- Improved retry, reconnection, and recovery behavior to prevent idle maintenance from blocking active work.
- Unified syntax highlighting across code blocks, source previews, and read-only diffs with copy and plain-text fallback support.
1 release this month · summarised from the official release notes · all months
Get started with nanobot
| PyPI | pip install nanobot-ai |
|---|
These commands use the packages tracked on this page. The vendor may recommend a different installer; see the official documentation.
nanobot alternatives
Best nanobot alternatives compared · All general-purpose AI agents
| # | Agent | Gid Score | Downloads 7d | 7d | 30d | Stars | Latest release | Price | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 8 |
OpenClawOpenClaw Foundation (nonprofit); created by Peter Steinberger |
70 | 2.8M | down 8.7% | down 27.1% | 395K+123/day | 2026.11.62d ago | Free (OSS) | |||
| 11 |
DeepSeek HarnessDeepSeek |
67 | 676K | down 14.9% | up 59.0% | 285K+622/day | — | Free | |||
| 14 |
Hermes AgentNous Research |
63 | 56.8K | down 26.1% | — | 265K+297/day | 0.23.3yesterday | Free (OSS) | |||
| 42 |
ManusButterfly Effect Pte. Ltd. (Manus AI), Singapore |
50 | — | — | — | — | — | $20/mo | |||
| 58 |
Cloudflare OSCloudflare |
46 | — | — | — | 11.8K+9/day | — | Free (OSS) | |||
| 63 |
|
452/4 | — | — | — | — | — | Free | |||
Head to head: nanobot vs DeepSeek Harness · nanobot vs Hermes Agent · nanobot vs Manus · nanobot vs OpenClaw
Tools used with nanobot
What 17 public GitHub projects that depend on nanobot use with it, from their dependency files.
- PyYAML58.8%
- Pydantic41.2%
- python-dotenv41.2%
- pytest41.2%
- Uvicorn35.3%
- Model Context Protocol SDK (Python)35.3%
- NumPy35.3%
- pandas35.3%
- Requests35.3%
- OpenAI Python Library35.3%
Used far more often here than in AI-built projects overall: what sets nanobot projects apart.
- json5 (Python)17.6% · 184.0×
- DingTalk Stream SDK17.6% · 138.0×
- ml_dtypes11.8% · 122.6×
- Slack SDK17.6% · 110.4×
- ModelScope11.8% · 73.6×
- json-repair17.6% · 61.3×
- websocket-client17.6% · 61.3×
- DDGS17.6% · 58.1×
Ratings from people who use nanobot
No ratings yet. Used nanobot? Rate it and say what it is good and bad at.
nanobot news
-
nanobot downloads up 28% in a week
23K PyPI downloads in the 7 days to Nov 19, against 18K the week before.
-
nanobot passes 50K GitHub stars
What changed: recent nanobot releases
3 stable releases in the last 90 days · Monthly summary · Full changelog
- One update flow, with your confirmation — Check releases in Settings → About, choose whether to receive Preview updates, review the exact version, and confirm installation and restart.
- Recovery that protects current data — Installer-managed updates can attempt repair and, when safe, restore the original program and service.
- Durable conversation history — SQLite brings model history and WebUI display history under gateway ownership. Existing JSONL history is imported with source backups retained.
- A clearer browser workbench — Navigate the full prompt history with bounded loading, search loaded messages and session handles, rename topics inline, and archive topics in batches.
- Your workbench, on the right host — Connect to existing remote instances with explicit host identity, pairing and reconnect feedback.
- Richer files and a better small-screen experience — Local-file and video attachments, image thumbnails, file cards, Markdown source views, reading-size controls, and touch and theme improvements.
Earlier releases
- A native terminal workbench — Start with nanobot, switch sessions, paste supported clipboard images, complete app and skill references, inspect context and file diffs, queue follow-ups, and use…
- Browser conversations side by side — Arrange up to four topics in resizable groups, keep sidebar ordering, drag saved sessions into mentions, and use temporary chats when you do not want a…
- Automations that are easier to manage — Switch between task and calendar views, inspect run replies, and adjust schedules with clearer controls. Editing a job preserves pending executions.
- Visible context and steadier long sessions — Per-round token/cache charts, context-compaction progress, paged history search, cumulative summaries, and file-read deduplication scoped to the actual…
- Apps and skills with clearer identities — SkillHub discovery, portable Agent Plugins, remote MCP OAuth and token refresh, visible connection failures, and consistent app logos and brand names in chat.
- A better small-screen experience — Home-screen installation through PWA support, narrower composer controls, draggable context sheets, touch-friendly chart details, and settings navigation that…
- One command to the workbench - nanobot webui prepares the local channel, starts the gateway, opens the browser, and leads first-time users into guided setup.
- A stronger browser workbench - Topic-style chats, composer model switching, quoted follow-ups, document attachments, native folder selection, responsive settings, and clearer fallback state.
- Agent coordination without core sprawl - Inline subagents can consult inside the active task, authorized work can continue through verification, and typed runtime context keeps provider-owned…
- Explicit long-running goals - Sustained execution is gated behind /goal, with clearer permission, cancellation, timeout, heartbeat, and local-trigger behavior.
- Files and tool activity you can inspect - Syntax-highlighted previews, live file-edit diffs, better code-copy behavior, structured tool errors, richer MCP artifacts, and preserved multimodal tool…
- A more operational runtime - Live configuration reloads, bounded session and exec state, safer restart/shutdown, cleaner cancellation ownership, more reliable scheduled work, and faster CI.
- More durable WebUI sessions — Segmented transcript storage keeps large conversations readable without throwing away old turns, fork/replay paths preserve assistant replies more reliably, session…
- Python SDK as a first-class runtime — The SDK now has safer concurrent run() behavior, explicit runtime controls, cleaner per-run hook handling, better MCP cleanup, improved facade lifecycle…
- Gateway and onboarding polish — The gateway gained background/service controls, cleaner foreground shutdown, safer Ctrl+C behavior, and better MCP stdio cleanup.
- Search, speech, and provider coverage — Web search now covers Firecrawl, Exa, Bocha, Keenable, and Volcengine paths.
- Reliability hardening across the agent core — Recent-history context is token-capped, idle auto-compact is on by default, Dream explains empty runs, malformed history is skipped, history cursors…
- Channels that behave more like real apps — Telegram gained richer message support with safer fallback detection.
- WebUI as a daily workbench — The WebUI is no longer just a packaged chat surface.
- Long-running agent stability — Goal mode and multi-turn work got sturdier: sustained goals keep the runner alive, goal iteration budget is extended, process_direct now has per-session locking…
- CLI Apps, MCP, and extension workflows — CLI Apps landed as a way to connect CLI-native tools, then got unified with MCP concepts through preset setup, capability mentions, stale install recovery…
- More providers, richer model routing — The provider surface widened with Ant Ling, Skywork, Novita, Step Plan, OpenAI API type selection, extraBody, and multiple image-generation providers…
- Channels and deployment polish — Signal support, Telegram webhook mode, Discord model slash commands, Docker/WebUI deployment fixes, and clearer secret/deployment docs make nanobot easier to run…
- Security and reliability hardening — WebSocket token issuance now requires auth, redirect targets are validated before web fetches, IPv6-mapped IPv4 addresses are normalized in SSRF checks, Teams…
nanobot prices over time
nanobot popularity and usage data
nanobot was downloaded 23K times from PyPI in the week ending Nov 19, 2026, up 27.7% on the week before and up 12.9% over the last 30 days compared with the 30 before. The GitHub repository has 50.2K stars, 1,165 of them added in the last 30 days. The latest stable release is 0.4.1, published Nov 6, 2026; there were 2 stable releases in the past 30 days.
Download the raw daily series: nanobot.csv
Gid Score breakdown
Gid Score shows public popularity and activity (usage, GitHub stars, mentions, growth), not the quality of the product. Each component is scored 0 to 100 from public signals; a dash means no data for it. How the Gid Score works.
How fresh these numbers are
| Number | Latest reading | Last checked |
|---|---|---|
| Package downloads | Nov 19, 2026 | Nov 21, 06:24 UTC |
| GitHub stars | Nov 21, 2026 | Nov 21, 06:26 UTC |
| Website rank (Tranco) | Nov 20, 2026 | Nov 21, 06:32 UTC |
We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.
nanobot FAQ
How much does nanobot cost?
Free MIT-licensed self-hosted software; users supply their own model providers or local models.
Is nanobot open source?
Yes. nanobot is open source, released under the MIT license.
How popular is nanobot?
nanobot was downloaded 23K times from PyPI in the week ending Nov 19, 2026, up 27.7% on the week before and up 12.9% over the last 30 days compared with the 30 before. The GitHub repository has 50.2K stars, 1,165 of them added in the last 30 days. The latest stable release is 0.4.1, published Nov 6, 2026; there were 2 stable releases in the past 30 days.
What is the latest version of nanobot?
The latest stable release we track is 0.4.1, published on Nov 6, 2026.
What are the alternatives to nanobot?
The closest alternatives in the same category by Gid Score are OpenClaw, DeepSeek Harness, Hermes Agent, Manus.
For the makers of nanobot badges, corrections
Add a live badge to your README or site:
[](https://agentgid.com/agents/nanobot/)
[](https://agentgid.com/agents/nanobot/)
[](https://agentgid.com/agents/nanobot/)
Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.
Where these numbers come from
PyPI: nanobot-ai. GitHub: hkuds/nanobot. Site rank: nanobot.wiki on the Tranco list. See data sources for how each one is collected.
Product facts and pricing were checked against: github.com, pypi.org.