ZeroClaw
ZeroClaw is a dual-licensed MIT OR Apache-2.0 AI assistant framework written in Rust for self-hosted, edge, and infrastructure use. Its architecture defines swappable provider, channel, tool, memory, runtime, and security-policy components through Rust traits. The project connects model providers to command-line and messaging interfaces, while adding memory backends, scheduled work, tools, and controls such as sender pairing, allowlists, encrypted secrets, filesystem scoping, and tunnel-oriented remote exposure.
Current releases provide builds for macOS, Linux on several architectures, Android ARM64, Windows x64 in at least recent release lines, and Linux package formats. The framework is free and open source; users still pay any model, hosting, messaging, tunnel, or external integration charges that apply.
Top features
- Trait-based components: providers, channels, tools, memory, runtimes, and security policies can be replaced behind defined interfaces.
- Provider routing: built-in and OpenAI-compatible adapters support hosted or local model endpoints without fixing the framework to one vendor.
- Messaging and API channels: documented options include CLI, Telegram, Discord, Slack, Matrix, Signal, WhatsApp, email, and webhooks.
- Memory backends: the architecture lists SQLite, PostgreSQL, Markdown, optional external systems, and a no-memory mode.
- Security controls: pairing, channel allowlists, encrypted secrets, scoped filesystem access, sandbox options, and restrictions on public binds are documented.
- Native or Docker runtime: operators can run directly or use a sandboxed container-oriented adapter.
- Cross-platform releases: archives and packages target desktop, server, mobile, and small-board architectures.
Use cases
- Run an assistant gateway on a small server, edge computer, robot controller, or field device.
- Build a messaging bot that can switch model providers and retain local memory.
- Add a custom provider, channel, tool, or policy implementation in Rust behind an existing trait.
- Deploy a self-hosted assistant with explicit sender pairing and egress or filesystem restrictions.
- Validate resource use and platform behavior on the intended target hardware before adoption.
No linked agents yet.
Register an agentGrok Build 1.0 Is xAI’s Bid to Make Terminal Coding Agents Usable Every Day
Grok Build 1.0.0 arrived on August 7, 2026. Its headline is not a new model or benchmark score, but the less glamorous work of making an agent that edits code, runs commands, and keeps working easier to control and recover.
AI Tools · Aug 8, 2026
Cloudflare Built a Browser for Agents, Not Humans, and Put It on Workers
Kitesurf is a Chromium alternative for agent workloads: Rust/Wasm pieces, V8 isolates, CDP compatibility, 3–7× less CPU/memory on Cloudflare’s corpus, and an honest list of things it still cannot do.
ai-agents · Aug 7, 2026
ChatGPT Puts GPT-5.6 Sol on Instant, Unlimited Luna on Free, and Leaves Work and Codex Alone
OpenAI’s August 6, 2026 ChatGPT update unifies paid Instant and reasoning on a retuned Sol, moves Free and Go to Luna with a later unlimited-text window, and keeps agent surfaces on the July models.
ai-models · Aug 6, 2026

