# NanoClaw Personal messaging agent with separate container workspaces. Canonical profile: https://ki-agentenkompass.de/en/agents/nanoclaw/ Editor: [Michael Meierhoff](https://ki-agentenkompass.de/en/about/) Sources reviewed: 2026-10-16 Basis: provider documentation. Examples are editorial suggestions, not hands-on product tests. ## Overview NanoClaw is listed as a separate product profile. - Provider: NanoCo - Product family: NanoClaw - Profile type: Self-hosted personal agent - Environment: Own computer or server - Connections: Tools and channels according to configuration - Setup: Set up Docker, Node.js and model access. - Pricing model: Open source; model and operating costs are separate. ## Key features ### Connect messaging Connect WhatsApp, Telegram, Slack and Discord through skills. Source: https://github.com/nanocoai/nanoclaw ### Schedule tasks Run recurring tasks on a schedule. Source: https://github.com/nanocoai/nanoclaw ### Separate workspaces Agent groups have their own containers and files. Source: https://github.com/nanocoai/nanoclaw ### Retain context Use per-group memory for follow-up tasks. Source: https://github.com/nanocoai/nanoclaw ### Choose models Connect additional model providers through extensions. Source: https://github.com/nanocoai/nanoclaw ## Example workflows ### Request a briefing through messaging - Your input: A topic and three approved web sources. - Possible result: A summary with sources in the connected channel. - What to check: Check content, permissions and actions taken. ### Separate two workspaces - Your input: Two agent groups with different sample files. - Possible result: Separate tasks with traceable file access. - What to check: Record time spent, corrections and costs. ## Requirements and limits ### Setup Requirement: Configure the model, channels and permissions yourself. Source: https://github.com/nanocoai/nanoclaw ### German in everyday use Not yet confirmed: End-to-end German interaction and output quality have not been tested here. Source: https://github.com/nanocoai/nanoclaw ### Reliability Not yet confirmed: Success rates for specific workflows have not been measured here. Source: https://github.com/nanocoai/nanoclaw ## A good fit if … - you can run Docker, Node.js and pnpm on your own machine or server - you want agents isolated in containers without holding raw credentials - you want to connect channels such as WhatsApp, Telegram or Slack yourself through skills ## Less suitable if … - you expect a monitoring dashboard, as the repository lists none - you do not want to run your own infrastructure, as you install and operate Docker, Node.js and pnpm yourself Editorial assessment based on provider documentation. Not a product test. ## Usable AI models Open choice By default NanoClaw runs Claude through the Claude Agent SDK. Bundled skills add Codex, Cursor, OpenCode with OpenRouter, Google or DeepSeek, and local models through Ollama, set per agent group. The README does not list arbitrary OpenAI-compatible endpoints as a separate route. - Model families named: Anthropic (Claude), OpenAI (GPT), Google (Gemini), DeepSeek · Other models can be connected freely - Model choice: You choose - Your own API key: Yes - Router or custom endpoint: Yes - Local models: Yes Documented · Sources accessed on: 2026-11-05 - [NanoClaw: README](https://github.com/nanocoai/nanoclaw) ## 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: 2026-10-19 ### Self-hosting with own model access With requirements NanoClaw installs its own container runtime and uses Anthropic credentials by default. Local Ollama models are documented as an alternative extension; check selected models and chat channels separately. 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. Official source: https://raw.githubusercontent.com/nanocoai/nanoclaw/main/README.md ## 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 NanoClaw (open source, MIT) with Docker containers and messenger channels Sources reviewed: 2026-11-05 ### Account used Documented: Channels such as WhatsApp, Telegram or Slack are added through skills. Agents hold no real API keys; the credential vault's gateway injects them into outgoing requests. Evidence: https://github.com/nanocoai/nanoclaw Evidence: https://docs.nanoclaw.dev/concepts/security ### Admin approval Documented with conditions: Approvals run through cards sent to admins by direct message, for example for credentials, self-modification and unknown chats. If no approver is reachable, the request is denied automatically. Evidence: https://docs.nanoclaw.dev/concepts/security ### Data access & team sharing Documented: An agent sees only the session folder, the group workspace and read-only shared code. Without an allowlist there are no extra mounts. All sessions of an agent group share the workspace, so separate trust domains belong in separate groups. Evidence: https://docs.nanoclaw.dev/concepts/security Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/isolation-model.md ## Prices and included usage Pricing checked: 2026-10-16 Open source; model and operating costs are separate. Total cost depends on access, usage and configuration. Pricing details: https://github.com/nanocoai/nanoclaw ## 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: NanoClaw on your own machine or server, agent groups in containers, scheduled tasks via the CLI Sources reviewed: 2026-11-05 ### Files and working state Documented: An agent group shares workspace, memory and CLAUDE.md across its sessions. Anything a scheduled task must keep between runs belongs in the group workspace under /workspace/agent. Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/scheduled-tasks.md Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/isolation-model.md ### Sign-ins Open in reviewed sources: The reviewed pages do not say how long channel connections remain valid. They only state that the container does not start if the credential vault is unreachable. Evidence: https://docs.nanoclaw.dev/concepts/security ### Starting further runs Documented: Tasks run once by timestamp or recurrently by cron in the installation's timezone, optionally with a script gate. An ungated recurring task that would fire more than four times in 24 hours is rejected. Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/scheduled-tasks.md ### Human intervention Documented with conditions: Approvals exist for specific actions (vault credentials, self-modification, unknown chats, optionally agent-to-agent messages). Not every action is confirmed individually; without a policy, agent-to-agent messages flow freely. Evidence: https://docs.nanoclaw.dev/concepts/security ### Detecting errors Documented: According to the repository there is no monitoring dashboard or debugging UI. For tasks, a CLI command shows run counts, failures and recent log entries, and the agent is asked to log after each run. Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/scheduled-tasks.md Evidence: https://github.com/nanocoai/nanoclaw ### Resuming after failures Documented with conditions: Failed recurring runs are delayed in steps from 2 to 60 minutes; after eight consecutive failures the series pauses until you fix and resume it. Retrying a single run is not described. If a setup step fails, Claude Code is invoked automatically to diagnose and resume from there. Evidence: https://github.com/nanocoai/nanoclaw/blob/main/docs/scheduled-tasks.md Evidence: https://github.com/nanocoai/nanoclaw ### Editorial assessment You run Docker, updates and troubleshooting yourself, though approvals and permissions are documented in detail. You need to check scheduled-task failures actively via the CLI. ## Memory and conversation continuity Scope: NanoClaw V2: agent groups and channel mapping Sources accessed: 2026-11-04 Shared agent memory and shared conversation can be configured separately. ### Storage and scope Documented: Agent groups own a workspace, CLAUDE.md and persistent memory under groups/; separate groups have their own directories and containers. - [NanoClaw repository](https://github.com/nanocoai/nanoclaw) - [Channel isolation model](https://raw.githubusercontent.com/nanocoai/nanoclaw/main/docs/isolation-model.md) ### Selection and timing Partly documented: Agents can save information when asked to remember. Exact automatic selection rules and write timing are unspecified here. - [Channel isolation model](https://raw.githubusercontent.com/nanocoai/nanoclaw/main/docs/isolation-model.md) ### Inspect, edit and delete Partly documented: Local workspace and CLAUDE.md files are accessible; a dedicated interface for individual memory facts is undocumented. - [NanoClaw repository](https://github.com/nanocoai/nanoclaw) ### Agent and channel boundaries Documented: Channels can share session and memory, share only memory, or belong to fully separate agent groups. - [Channel isolation model](https://raw.githubusercontent.com/nanocoai/nanoclaw/main/docs/isolation-model.md) ### Retrieval Partly documented: Saved group memory can be recalled in other associated sessions; the search algorithm is not explained here. - [Channel isolation model](https://raw.githubusercontent.com/nanocoai/nanoclaw/main/docs/isolation-model.md) ### Session reset and compaction Partly documented: Separate conversations of the same agent group retain shared memory. Specific reset and compaction behavior remains unspecified in these sources. - [Channel isolation model](https://raw.githubusercontent.com/nanocoai/nanoclaw/main/docs/isolation-model.md) ## Context Configure the model, channels and permissions yourself. Provider sources reviewed; examples are editorial. No hands-on tests. ## Open points Requirements and limits, not yet confirmed: - German in everyday use - Reliability ## Sources - https://github.com/nanocoai/nanoclaw ## Provider Publishes NanoClaw and the linked product information. - Operator: NanoCo - Organization: Open-source project - Website: https://github.com/nanocoai/nanoclaw - Reviewed: 2026-10-16 - Provider source: https://github.com/nanocoai/nanoclaw [Sources and method](https://ki-agentenkompass.de/en/sources/)