Skip to content

v0.1.0 — Initial public release

Choose a tag to compare

View all tags
@a-radwan-20 a-radwan-20 released this 31 May 14:07
· 80 commits to main since this release
v0.1.0
3995daa

First public beta

AI Butler is a self-hosted personal AI agent with exceptional memory, any AI
model, a built-in web UI, and a 12-channel messaging interface. This is the
first public beta — the core is production-ready, several advanced features
are in beta and labeled clearly.

Core stats:

  • 1,586 tests passing (race-free, go test -race -count=1)
  • 114 Go packages
  • 67 SQLite tables, 19 migrations
  • 59 internal security audit passes, 74 findings fixed
  • 0 CVEs (govulncheck verified on every CI run)
  • ~10 direct dependencies, zero CGO
  • Apache 2.0 licensed with explicit patent grant

Production-ready features

Everything in this list is tested end-to-end and ready for daily use.

Core runtime

  • Agent loop with streaming responses, tool-call orchestration, and
    capability-gated execution
  • Memory system combining FTS5 full-text search, an entity knowledge graph,
    vector embeddings, and hybrid ranking via Reciprocal Rank Fusion
  • Per-project BUTLER.md instruction files (like .cursorrules)
  • Three-tier config system (Settings / Configurations / Options) with the
    Butler Three Enriches pattern
  • Single Go binary, zero CGO, cross-compiles to Linux/macOS/Windows/FreeBSD
    on amd64/arm64/arm/riscv64

AI providers

  • Anthropic Claude — direct API integration, streaming, tool use
  • Ollama — auto-detects running instance, uses it for both chat and
    memory embeddings; fully local option with zero cloud calls

User interfaces

  • Web chat — 6-panel sidebar dashboard (Chat · Home · Memories ·
    Connected Apps · Spending · Settings). Welcome screen with 6 starter
    prompts. Live budget pill. Connection status. Mobile-responsive drawer.
    Auto/Light/Dark theme. File upload + voice upload. English + Arabic RTL.
  • Terminal REPL — interactive streaming, slash commands, session resume
  • Dashboard API — 16+ JSON endpoints (/api/dashboard/*) for stats,
    memory, agents, costs, swarm topology, capabilities, IoT, plugins,
    transactions, AI providers, audit log, registry, agent-card

Developer tools

  • file.read, file.write, file.edit, file.list, file.search —
    workspace-bounded with path enforcement
  • shell.exec — allowlisted commands, sandboxed via unshare (Linux)
    or sandbox-exec (macOS)
  • git.status, git.diff, git.commit, git.log, git.branch,
    git.pr_create — wrapping the real git CLI

Operations

  • Scheduler — natural language → cron, delivered to any channel,
    persistent across restarts
  • Cost tracking — per-model, per-session, live dashboard, alerts
  • MCP client — auto-connects to external MCP servers on boot
  • Vault — credential storage in OS keyring (Keychain, GNOME Keyring,
    Windows Credential Manager) with age-encryption fallback
  • Capability engine — per-tool grants + tamper-evident audit trail
  • Integrity checker — aibutler integrity runs SQLite PRAGMA integrity_check + migration validation + orphan row detection

Deployment

  • Docker + Docker Compose (standalone, + ollama, + full stack)
  • Kubernetes via Helm chart (deploy/helm/)
  • systemd service (deploy/systemd/)
  • Raspberry Pi (ARM64) — first-class target

Beta features — code complete, needs real-world validation

All of these compile, pass unit tests, and work in isolation. We're
asking the community to help validate them end-to-end with real
credentials, real third-party services, and real usage.

Messaging channels (10 beta)

  • Telegram (Bot API with long polling + webhook mode)
  • Slack (Bolt SDK with Socket Mode)
  • Discord (Gateway)
  • WhatsApp (Meta Cloud API)
  • Microsoft Teams (Bot Framework)
  • Google Chat (Cards v2)
  • LINE (Messaging API with Flex Messages)
  • IRC
  • Custom webhook (HTTP)
  • Nostr (NIP-04 direct messages)

AI providers (5 beta)

  • OpenAI GPT + Azure OpenAI
  • Google Gemini
  • xAI Grok
  • LM Studio / vLLM / Groq / DeepSeek (OpenAI-compatible)

Advanced

  • Multi-agent swarm orchestration — engine works, first cookbook
    example coming in v0.2
  • Full Google A2A v2 protocol — handler is spec-compliant, formal
    conformance suite pending
  • MCP server mode — expose Butler's tools to other MCP clients
    (Claude Desktop, etc.)
  • WASM plugin sandbox via Extism — runtime
    ready, sample plugins coming in v0.2
  • OIDC SSO (Auth0, Okta, Keycloak, Authelia, ZITADEL)
  • FIDO2 / WebAuthn hardware security keys
  • TOTP 2FA (RFC 6238, any authenticator app)
  • RBAC with admin/user/viewer/agent roles
  • LAN mode with mDNS discovery + PIN pairing
  • Subprocess bridges for wrapping CLI tools as first-class Butler tools
  • Smart home via Home Assistant — tool surface + PIN safety gating
    ready, HA adapter in final wiring
  • Whisper STT (local binary or cloud API)
  • Piper TTS (local, CPU-only)

Roadmap for v0.2

  • Hosted playground at demo.aibutler.dev (rate-limited, Haiku-only,
    read-only sandbox)
  • Smart home: working Home Assistant adapter out of the box
  • Plugin marketplace with sample plugins
  • Multi-agent swarm cookbook with real-world delegation examples
  • Voice TUI mode (terminal mic capture + playback)
  • Image generation adapters (Flux, Stable Diffusion, DALL-E)
  • ElevenLabs TTS adapter
  • Internet access mode with autocert TLS + password + TOTP
  • PWA (installable web app)
  • Homebrew formula
  • SLSA Level 3 provenance + cosign binary signing
  • 14-language i18n
  • External third-party security audit

Installation

# Build from source (recommended)
git clone https://github.com/LumabyteCo/aibutler.git
cd aibutler && CGO_ENABLED=0 go build -o aibutler .

# Or download the release binary for your platform
curl -sSL https://github.com/LumabyteCo/aibutler/releases/download/v0.1.0/aibutler_0.1.0_Linux_x86_64.tar.gz | tar xz
sudo mv aibutler /usr/local/bin/

# Or Docker
docker pull ghcr.io/lumabyteco/aibutler:v0.1.0

Quick start

./aibutler vault set anthropic_api_key sk-ant-...
./aibutler run
# Open http://localhost:3377

Full walkthrough with screenshots: docs.aibutler.dev/getting-started/quick-start/

Breaking changes

None — this is the initial release.

Acknowledgements

AI Butler is built on the shoulders of:

Contributors

This release was built by LumaByte Co. Community
contributions welcome — see CONTRIBUTING.md for the full
guide on how to push beta features to ready, write sample plugins, test
channels, and help shape the roadmap.

Assets 2
Loading