Cameleer

Cameleer packages AI agent management into a desktop app modeled on internal enterprise tooling: an agent directory for creating and assigning agents with specific roles and personas, a Kanban/backlog system for tracking tasks assigned to those agents, workspace chat for talking directly to assigned agents, and a runtime/audit view showing live run logs, pending tool approvals, and complete audit history.

Built as a Tauri application (Rust backend, React/TypeScript frontend, SQLite for local state), it’s local-first by design, storing entity data in a local_state.db file managed by Rust rather than depending on a cloud backend. The frontend and backend communicate exclusively through Tauri’s invoke command bridge, with Zustand handling UI-layout state separately from the SQLite-backed entity data.

As of the reviewed v0.1 production candidate (verified on macOS/Apple Silicon), the project reports 44/44 backend tests passing including an end-to-end smoke test. No LICENSE file is present in the repository, so despite being publicly available on GitHub, it currently defaults to standard copyright rather than a formal open-source license.

What You Get

• An agent directory for creating, managing, and assigning AI agents with specific roles and personas
• Kanban and backlog-based task tracking for work assigned to agents
• Workspace chat for talking directly to assigned agents
• A runtime/audit view showing live agent run logs, pending tool approvals, and complete audit history

Common Use Cases

• Managing a small team of AI agents with distinct roles and responsibilities within one workspace
• Tracking agent-assigned tasks through a Kanban board instead of ad hoc chat threads
• Reviewing an audit trail of what agents did and which tool calls required approval
• Running a local-first agent management workflow without depending on a cloud service

Under The Hood

Architecture Cameleer strictly separates the React/TypeScript frontend from the Rust/SQLite backend, communicating exclusively through Tauri’s invoke command bridge (wrapped in src/api/) rather than a general-purpose API layer — a deliberate boundary that keeps UI state (managed by Zustand) separate from entity data (fetched dynamically from the Rust-managed local_state.db SQLite file). This domain separation means the frontend has no direct database access, all persistence logic living in the Rust layer.

Tech Stack Tauri as the application shell, React/TypeScript for the frontend, Rust for the backend, and SQLite (via local_state.db) for local persistence — a fully local-first stack with no cloud dependency by default. Zustand handles client-side UI state.

Code Quality The project reports 44/44 backend tests passing, including an end-to-end smoke test, for its v0.1 production candidate verified on macOS/Apple Silicon — concrete, checkable testing claims rather than vague assurances, though the absence of a LICENSE file and limited community engagement reflect its very early stage.

What Makes It Unique Rather than being a chat interface with agent branding, Cameleer models agent management on genuine enterprise tooling patterns — role-based agent directory, Kanban task assignment, and a formal audit/approval trail for tool calls — treating agents as workspace members with accountability rather than a single conversational assistant.

Self-Hosting

Licensing Model No LICENSE file is present in the repository. Despite being publicly viewable on GitHub, this means the code defaults to standard copyright (all rights reserved) rather than a formal open-source license — treat it as source-available rather than freely licensed until a LICENSE file is added.

Self-Hosting Restrictions Not applicable; it’s a local desktop app with no hosted service.

License Key Required No.