AI ASSISTANTS • CURATED • UPDATED AUG 20, 2026

NanoClaw

Minimalist, container-isolated personal AI agent framework
Price
Free
API
No
Weights
Yes

NanoClaw is an open-source personal AI agent runtime built by Gavriel Cohen as a smaller, auditable alternative to OpenClaw. Each agent session runs in its own Docker container with scoped permissions and self-destructs when the task ends. In August 2026 it added a Slack integration that lets you provision persistent AI agent teams and colleagues from a single message, running on customer infrastructure.

The agent landscape is polarised between all-in-one platforms with huge codebases and small, custom rigs. NanoClaw occupies the small, auditable end: roughly 500 lines of TypeScript, container isolation by default, and a fork-and-own model that makes the agent's capabilities explicit rather than hidden behind plugins.

Pricing Free
Platforms local, slack
API No
Open Source Yes
Modalities AI Assistants
Best For Best for Auditable Agents
Date Added 2026-08-20
  • ✓Very small codebase, easy to audit
  • ✓Container isolation for every session
  • ✓Scoped permissions so one agent cannot see another channel
  • ✓Slack integration for persistent agent teams
  • ✓Self-hosted, with low running cost
  • ⚠Fork-and-customize model requires technical setup
  • ⚠Fewer out-of-the-box integrations than OpenClaw
  • ⚠Primarily designed around Claude today
  • ⚠Smaller community ecosystem than larger agent platforms
  • ⚠Self-hosting means you manage infrastructure and secrets

Fork the NanoClaw repository on GitHub, customise it to your environment, and run it locally or in a container. Add the Slack skill to provision agent teams from a single Slack message. Start with a single channel and scoped permissions before expanding.

1 Scope each agent to one channel or task surface at first
2 Audit the skill code before adding new capabilities
3 Use containers to isolate sessions from each other and from host data
4 Compare total cost against OpenClaw if you need many integrations
5 Document what each persistent agent is allowed to touch

Persistent Slack Agent Team

Create a team of agents that live in a project channel and handle recurring tasks.

STEPS:
  1. Install the NanoClaw Slack skill
  2. Provision the team from one message in the channel
  3. Assign each agent a role and permission scope
  4. Connect the agents to the tools they need
  5. Review logs and revoke access when the project ends
Free Completely free
View NanoClaw Alternatives (2026) →

Compare NanoClaw with 5+ similar ai assistants AI tools.

❓
Q

Is NanoClaw free?

A

Yes, NanoClaw is completely free to use.

Q

Does NanoClaw have an API?

A

No, NanoClaw does not appear to offer a public API.

Q

What is NanoClaw best for?

A

NanoClaw is best for Auditable Agents. NanoClaw is an open-source personal AI agent runtime built by Gavriel Cohen as a smaller, auditable alternative to OpenClaw. The agent landscape is polarised between all-in-one platforms with huge codebases and small, custom rigs. NanoClaw occupies the small, auditable end: roughly 500 lines of TypeScript, container isolation by default, and a fork-and-own model that makes the agent's capabilities explicit rather than hidden behind plugins.

Q

What platforms does NanoClaw support?

A

NanoClaw supports local, slack.

Q

Is NanoClaw open source?

A

Yes, NanoClaw is open source. You can access the source code on GitHub at https://github.com/nanocoai/nanoclaw.

Q

How do I get started with NanoClaw?

A

Fork the NanoClaw repository on GitHub, customise it to your environment, and run it locally or in a container. Add the Slack skill to provision agent teams from a single Slack message. Start with a single channel and scoped permissions before expanding.

🏷️

Work on NanoClaw? You're hand-reviewed in our directory. Add this badge to your site — it links back to this profile.

Featured on CuratedAI