CueMarshal – AI-Powered DevOps for Software Teams

From issue to reviewed PR without leaving Git.

CueMarshal is self-hosted AI DevOps: GitHub Actions + an AI engineering orchestra on your own infrastructure. Open an issue, and CueMarshal plans, builds, reviews, tests, and documents the change through Git-native branches and pull requests. You keep final control: review every PR, request edits, and merge when ready.

• Marshal (Conductor) — routes work and keeps execution moving
• Ava (Architect) — designs the solution and opens a spec PR
• Dave (Developer) — writes the implementation on a feature branch
• Reese (Reviewer) — checks for bugs, risks, and code quality issues
• Tess (Tester) — runs test suites and validates behavior
• Devin (DevOps) — handles infrastructure and deployment tasks
• Dot (Docs) — keeps documentation clear and current
• Linton (Linter) — enforces code quality checks before merge

Every step lives in Gitea as issues, branches, pull requests, and labels. Full audit trail. Full human control.

⚠️ Early stage. Core workflow is functional. Self-improvement engine works but needs more testing. Mobile app and Kubernetes Helm charts are in progress.

Why CueMarshal?

Use Cases:

• Ship faster — Cut code review time from hours to minutes
• Better quality — AI reviewers catch security issues humans miss
• Data sovereignty — Your code never leaves your servers
• Cost control — ~$0.50/issue in API costs vs $100+/month in cloud CI

Quick Start · Architecture · Agents · Documentation

Meet the Orchestra

Each team member has a distinct avatar and working style, so activity feels like a real engineering squad instead of a black box.

Quick Start

# Clone the repository
git clone https://github.com/cuemarshal/cuemarshal.git
cd cuemarshal

# One-command setup (handles secrets, deps, and health checks)
./quick-start.sh

Open http://localhost:3300 — Gitea is ready.

Need full control? See the detailed setup guide.

Prerequisites

• Docker and Docker Compose v2+
• 8 GB RAM minimum (16 GB recommended)
• Local Ollama at http://localhost:11434 running gemma4:26b, or at least one cloud LLM API key:
  • Groq (free, fast — primary)
  • Google Gemini (free — fallback)
  • Azure AI (paid — second fallback)

What's Implemented

• ✅ Conductor orchestrator (task decomposition + agent routing)
• ✅ 7 specialized AI agents (architect, developer, reviewer, tester, devops, docs, linter)
• ✅ Git Flow execution (issue → branch → PR → review → merge)
• ✅ LiteLLM gateway with 3-provider fallback (Groq → Gemini → Azure)
• ✅ MCP servers for tool access (Gitea, Conductor, System)
• ✅ Self-improvement loop (system scans and improves its own codebase)
• ✅ Docker Compose deployment (11 services, one command)
• ⏳ Mobile app (React Native, in progress)
• ⏳ Kubernetes Helm charts (in progress)
• ⏳ Multi-repo support (planned)

How It Works

User creates issue in Gitea
         │
    Conductor receives webhook
         │
    Conductor decomposes task → assigns to agents
         │
    ┌────┴──────────────────────────────────┐
    │  Architect  →  spec PR                │
    │  Developer  →  implementation PR      │
    │  Reviewer   →  review comments + fix  │
    │  Tester     →  test results           │
    │  DevOps     →  infra/deployment PR    │
    │  Docs       →  documentation PR       │
    │  Linter     →  pre-merge quality gate │
    └────────────────────────────────────────┘
         │
    Human reviews + merges
         │
    Done. Full audit trail in Git.

Architecture

User (Mobile App / Gitea UI)
         │
    Conductor (TypeScript) ── Redis/BullMQ
         │          │
    MCP Servers    LLM Gateway (LiteLLM)
    ├── Gitea MCP      ├── Groq
    ├── Conductor MCP  ├── Google Gemini
    └── System MCP     └── Azure AI
         │
    Gitea Server ── PostgreSQL
         │
    Runners (Gitea Act Runner + OpenCode)
    ├── Developer Agent
    ├── Reviewer Agent
    ├── Tester Agent
    ├── Architect Agent
    ├── DevOps Agent
    ├── Docs Agent
    └── Linter Agent

See Architecture Overview for full diagrams and data flows.

Services

Who Should Use CueMarshal?

• Software teams wanting AI-assisted code review and task automation
• Engineering leaders concerned about vendor lock-in or cloud costs
• Organizations with data residency requirements (your code stays on your servers)
• Developers already using Gitea who want AI-powered workflows
• Builders who want to experiment with AI agent orchestration

Technology Stack

Documentation

Contributing

CueMarshal uses its own platform to manage contributions — improvements flow through Gitea issues and PRs, often executed by the AI agents themselves.

To contribute:

1. Fork the repository and create a feature branch from main
2. Open an issue describing what you plan to change before submitting a PR
3. Follow Git Flow: one logical change per branch, one PR per issue
4. Run tests before opening a PR:

   cd services/mcp-servers && npm test
   cd services/conductor && npm test
   

5. Ensure Docker Compose starts cleanly with ./quick-start.sh

PRs are reviewed by the CueMarshal reviewer agent. Humans have final merge authority.

License

MIT — see LICENSE for details.