What it does
5 features
nanobot · Self-hosted personal agent
Product and access
nanobot is listed as a separate product profile.
Core features
5 features from provider sources · checked on 16 October 2026. Availability and limits are listed below.
Usable AI models
Open choice
nanobot does not supply a model account: you connect your own keys or sign-ins, routers such as OpenRouter, or an OpenAI-compatible endpoint. Models run locally through Ollama, LM Studio or vLLM.
Model families named
- Anthropic (Claude)
- OpenAI (GPT)
- DeepSeek
- xAI (Grok)
- Alibaba (Qwen)
- Moonshot (Kimi)
- MiniMax
- Other models can be connected freely
- Model choice
- You choose
- Your own API key
- Yes
- Router or custom endpoint
- Yes
- Local models
- Yes
The rating covers the generally available main access. Documented · Sources accessed on
Does it fit my task?
2 practical tasks · 3 requirements and limits
A good fit if …
- you want to run a personal agent on your own machine or server
- you want to talk to it through chat channels such as Telegram, Discord or Slack
- you can handle the installation and model configuration yourself
Less suitable if …
- you want a managed service with central user and permission management
- you expect automatic resumption after a crash or restart
Editorial assessment based on provider documentation. Not a product test.
Practical tasks
Editorial examples based on documented features. No product tests performed.
Summarize a document
- Your input
- An anonymized PDF and a guiding question.
- Possible result
- A short summary for subject-matter review.
- What to check
- Check content, permissions and actions taken.
Trigger a task through chat
- Your input
- A configured channel and a bounded test task.
- Possible result
- A result in the connected chat channel.
- What to check
- Record time spent, corrections and costs.
Summarize a document.
Your possible prompt
Organize these shared documents by topic. Create an overview with deadlines and next steps. Prepare draft replies for my approval.
Adjust the task, data and permissions.
Requirements and limits
Can I use it?
With requirements
Europe & regional availability
Installation instructions and model prerequisites reviewed. The assessment covers the local operating path described; complete country and feature availability of all connected services remains open.
Reviewed scope: Local software and documented self-hosting; external services assessed separately
Reviewed on:
With requirements
Self-hosting with own model access
Python installation and a local or server-side gateway are documented. A supported model provider or your own model server is required; nanobot does not include a model account.
Region, reason and source
Region covered: Europe: own computer or server; model services are country-specific
Documented reason: Technical assessment based on installation and provider selection. These sources do not establish a legal cause for regional restrictions.
Accounts & permissions
Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use.
Configuration covered: Self-hosted nanobot instance with chat channels and the WebUI
Sources reviewed:
Show all 3 points
Open in reviewed sources
Account used
The project page lists channels such as Telegram, Discord and Slack but does not describe which identity (bot account, token) the agent uses there.
EvidenceDocumented with conditions
Admin approval
Check direct-message pairing, the allowlist and group-access policies separately for each chat platform: approving a private sender does not restrict group access on every platform. WebUI access grants administration of the same agent instance.
EvidenceDocumented with conditions
Data access & team sharing
For a selected project folder, the default permission is workspace-only access. One instance shares its agent profile, memory and skills across projects; separate instances are meant for cases where data or preferences must stay isolated.
Evidence
Memory and conversation continuity
5 of 6 criteria documented
Conversation compaction produces history; dreaming transfers selected insights into versioned files.
Current memory and dream system
Sources accessed:
Documentation by criterion
StorageDocumented
Storage and scope
Workspace files SOUL.md, USER.md and memory/MEMORY.md form durable memory; history.jsonl holds compacted conversation segments.
SelectionDocumented
Selection and timing
Compaction runs on context growth, inactivity or /compact; enabled dreaming processes new summaries on schedule or through /dream.
Edit & deleteDocumented
Inspect, edit and delete
Memory files can be inspected and corrected; GitStore versions changes. Archive cleanup is separate from durable facts.
Access boundariesDocumented
Agent and channel boundaries
One instance shares profile and memory across projects; separate instances create separate personal memories.
RetrievalPartly documented
Retrieval
Durable profile files supply context; dreaming reads new archive summaries. An additional vector index is not established here.
New conversationsDocumented
Session reset and compaction
Memory files survive new sessions. Compaction creates summaries; later archive cleanup does not remove original chat files.
Cost and effort
Prices as of 16 October 2026
Prices and included usage
Pricing checked
Open source; model and operating costs are separate.
- Total cost depends on access, usage and configuration.
Setup
Install the installer or Python package and configure a model.
Operation & oversight
Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured.
Deployment covered: Self-hosted gateway in background mode with scheduled automations
Sources reviewed:
Editorial assessment
You run and monitor the instance yourself. After restarts you resume work manually and check access rules per chat channel.
Show all 6 points
Documented
Files and working state
Files created by tools remain after the chat closes. Conversation history can be kept in SQLite; temporary chats stay out of saved history and long-term memory.
EvidenceOpen in reviewed sources
Sign-ins
The project page does not describe how long channel connections stay valid or what happens on expiry.
EvidenceDocumented
Starting further runs
Scheduled automations run via cron. Tasks can be edited or paused and recent runs inspected; in the WebUI you manage schedules under Tasks. Missed times are not replayed after a gateway restart.
EvidenceOpen in reviewed sources
Human intervention
The project page names no approval flow for tool actions, only permission settings such as the default of workspace-only access.
EvidenceDocumented with conditions
Detecting errors
The command nanobot gateway logs shows background gateway logs. Failure notifications are not described.
EvidenceDocumented
Resuming after failures
After a restart you reopen the topic, review saved results and explicitly resume interrupted work. Uncertain tool actions are not replayed and interrupted subagents are not relaunched automatically.
Evidence
Evidence and open questions
1 source · 2 open points
Open points
The reviewed sources contain no reliable information on these points. A gap says nothing about the product.
Requirements and limits: not yet confirmed
Sources and evidence
Provider sources reviewed; examples are editorial. No hands-on tests.
Editorial team: Michael Meierhoff · sources reviewed on . Provider information is not independent evidence of performance.
Provider
Publishes nanobot and the linked product information.
- Operator
- HKUDS
- Organization
- Open-source project
- Website
- github.com
Changes to this profile 1
Added nanobot with capabilities, sources, pricing and open European availability questions.
These notes describe editorial changes. Source and pricing review dates appear in their respective sections.