Blog
20 août 2026
NanoClaw brings on-demand agent teams to Slack's Add to Slack
NanoClaw is a launch partner for Slack's Add to Slack. Ask one agent for a team and it stands up a writer, a developer, and a designer — each with its own persona, memory, and sandbox — collaborating with you in the channel and in a shared Canvas.
13 août 2026
How templates work, and how to build one worth sharing
Agent Templates turn a working NanoClaw agent into something you can move: a plugin base in the open Agent Plugins standard plus a NanoClaw extension for persona and scheduled tasks. Here's what's inside one, what install does, and what makes a template worth sharing.
29 juillet 2026
NanoClaw Partners with Echo to Patch Known Vulnerabilities in the Agent's Runtime, Browser, and Libraries
NanoClaw and Echo are shipping a hardened agent runtime: the same NanoClaw environment, rebuilt so the software inside it carries near zero known vulnerabilities.
7 mai 2026
How to Deploy NanoClaw on exe.dev: A Step-by-Step Guide
Step-by-step guide to deploying NanoClaw on exe.dev — sign up, activate your subscription, spin up a VM, and run the NanoClaw setup.
23 avril 2026
NanoClaw V2: Human-in-the-loop approvals and Agent to Agent collaboration
NanoClaw V2 adds human-in-the-loop approvals through rich cards in messaging apps and persistent agent-to-agent communication, turning the safety/capability tradeoff into a dial you can tune.
24 mars 2026
NanoClaw Adopts OneCLI Agent Vault
Every NanoClaw agent will access external services through OneCLI's Agent Vault, a gateway that handles credential injection, access policies, and approvals so agents never hold raw API keys.
13 mars 2026
Run NanoClaw in Docker Sandboxes with One Command
Every agent gets its own isolated container inside a micro VM. No dedicated hardware needed. No complex setup.
3 mars 2026
NanoClaw n'a pas de fonctionnalités (et c'est tant mieux)
Au lieu de livrer des fonctionnalités, NanoClaw utilise des skills — des instructions qui apprennent à Claude comment modifier votre codebase. Le résultat : un logiciel ultra-personnalisé où chaque utilisateur dispose exactement des capacités dont il a besoin.
28 février 2026
Ne faites pas confiance aux agents IA
Les agents IA ont besoin d'un modèle de sécurité qui part du principe que les choses vont mal tourner. La bonne réponse n'est pas de meilleurs contrôles de permissions, mais une architecture qui rend la confiance inutile.