Jumbo CLI tackles "agent amnesia," inconsistent code quality, and vendor lock-in plaguing AI coding agents. It provides developers with persistent memory and context orchestration, ensuring continuity across sessions and improving AI-generated code reliability. Aimed at users of agents like Claude, Copilot, and Gemini, it boosts productivity by eliminating repetitive context-setting and ensuring production-ready code.

How It Works

Jumbo CLI captures and manages project context ("memory") for AI agents, using a Clean Screaming Architecture with Event Sourcing (JSONL logs) and CQRS (SQLite projections). This enables Jumbo to inject relevant project details—domain, solution, operational goals—into agent sessions, guiding them through refinement, implementation, and codification. Its harness- and model-agnostic design ensures context portability, allowing seamless switching between AI tools without losing project history.

Quick Start & Requirements

• Install: npm i -g jumbo-cli
• Run: jumbo in your project directory; the CLI guides setup.
• Prerequisites: AI agents supporting hooks or AGENTS.md (tested with Claude Code, Copilot, Gemini). VS Code with Copilot confirmed.
• Resource Footprint: Fully local, fast, private operation. Data stored in .jumbo/.

Highlighted Details

• Comprehensive Memory: Stores project domain, solution architecture, dependencies, decisions, guidelines, invariants, and operational goals/sessions.
• Model Agnostic: Integrates and allows interchangeable use of various AI coding agents and models.
• Local-First & Private: All data resides on the user's machine.
• Event Sourcing & CQRS: Utilizes robust patterns for auditable history and optimized read performance.

Maintenance & Community

Originates from a developer solving personal workflow issues. A cloud team version ("Jumbo Herd") is planned; sign-ups at jumbocontext.com/herd. No other community or contribution details are provided.

Licensing & Compatibility

• License: AGPL-3.0.
• Compatibility: AGPL-3.0 is a strong copyleft license, requiring derivative works to be open-sourced under the same license. This may restrict integration into proprietary, closed-source commercial applications.

Limitations & Caveats

• Team Collaboration: Currently individual-focused; team features depend on the upcoming cloud version. Default .gitignore for .jumbo/ suggests potential sharing challenges.
• Agent Integration: Full functionality requires agent hook/AGENTS.md support.
• AGPL-3.0 License: Copyleft nature may be a blocker for commercial, closed-source projects.