Labsco
ashlrai logo

ashlr-workbench

★ 2

from ashlrai

Local agent workbench bundling OpenHands, Goose, Aider, and ashlrcode against one local LLM, with ashlr-plugin MCP servers pre-wired.

🔥🔥✓ VerifiedFreeQuick setup
#devops#git

ashlr-workbench

Your local coding-agent HQ — four agents, one local LLM, ten MCP servers, zero cloud dependencies.

What this is

A curated, fully-local toolbox that wires four open-source coding agents (OpenHands, Goose, Aider, ashlrcode) to a single LM Studio model (Qwen3-Coder-30B) and the same set of ten ashlr-plugin MCP servers. The workbench supplies one CLI (aw), one healthcheck, and one update path so the whole stack feels cohesive instead of four loose tools you have to babysit. No data leaves the machine unless you opt into a cloud fallback.

The agent lineup

AgentFormBest forCost
OpenHandsDocker daemon + GUIAutonomous multi-step PRs, browser-driven tasksHeavy RAM
GooseNative Rust CLIFast tool-using sessions, smart-approve loopLight
AiderPython CLISurgical file-by-file refactors with explicit diffsLight
ashlrcodenpm/Bun CLI (Mason's)Personal day-to-day work, hooks-aware, cloud fallbackLight

Pick the one that matches the shape of the work — see "Usage examples" below.

Architecture

                ┌─────────────────────────────────────────────┐
                │              ashlr-workbench                │
                │                                             │
                │   bin/aw  →  scripts/{start,health,update}  │
                └─────────────────────────────────────────────┘
                                    │
        ┌───────────────────┬───────┴────────┬──────────────────┐
        ▼                   ▼                ▼                  ▼
   ┌──────────┐        ┌────────┐       ┌────────┐         ┌──────────┐
   │ OpenHands│        │ Goose  │       │ Aider  │         │ashlrcode │
   │ (docker) │        │ (rust) │       │(python)│         │ (bun)    │
   └────┬─────┘        └───┬────┘       └───┬────┘         └────┬─────┘
        │                  │                │                   │
        └──────────────┬───┴────────────────┴───────────────────┘
                       │  same MCP surface
                       ▼
            ┌─────────────────────────┐         ┌────────────────┐
            │  ashlr-plugin (10 MCPs) │   ←──   │  LM Studio     │
            │  efficiency / sql /     │         │  Qwen3-Coder   │
            │  bash / tree / http /   │         │  -30B  :1234   │
            │  diff / logs / genome / │         └────────────────┘
            │  orient / github        │
            └─────────────────────────┘         ┌────────────────┐
                                                │ Ollama :11434  │
                                                │  (fallback)    │
                                                └────────────────┘

All four agents talk to the same LLM and the same MCP tools, so behavior is consistent regardless of which one you launch.