PluginBench
MCP Server
Maintained
MIT

OpenVibeCoding Read-only MCP MCP Server

io.github.xiaojiou176-open/openvibecoding-readonly

Read-only MCP for monitoring OpenVibeCoding runs, workflows, proofs, queues, replays, and incidents.

What is the OpenVibeCoding Read-only MCP MCP server?

The OpenVibeCoding Read-only MCP server provides read-only access to OpenVibeCoding orchestrator data including runs, workflows, proofs, queues, replays, and incidents. It enables AI agents to monitor and inspect long-running coding tasks without modifying state.

This is a read-only MCP server that exposes monitoring and inspection capabilities for OpenVibeCoding, a long-haul AI coding orchestrator. Use it to query run status, workflow history, proof artifacts, queue state, replay logs, and incident records—useful for tracking AI-driven coding tasks and understanding execution history.

How to install OpenVibeCoding Read-only MCP

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "openvibecoding-readonly": {
      "command": "uvx",
      "args": [
        "openvibecoding-orchestrator"
      ]
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • runs — Query and inspect OpenVibeCoding runs
  • workflows — Access workflow definitions and execution history
  • proof — Retrieve proof of work artifacts
  • queue — Monitor task queue state
  • replay — Access execution replay logs
  • incidents — Query incident records and diagnostics

Use cases

  • Monitor long-running AI coding tasks and their current status
  • Review proof of work artifacts and execution history
  • Inspect workflow definitions and execution logs
  • Track task queue state and pending operations
  • Replay past executions to debug or audit coding runs
  • Query incident records for troubleshooting failed tasks

OpenVibeCoding Read-only MCP MCP server FAQ

What is the OpenVibeCoding Read-only MCP server?

It's a read-only MCP server that provides access to OpenVibeCoding orchestrator data: runs, workflows, proofs, queues, replays, and incidents. You can inspect and monitor long-running AI coding tasks without modifying state.

Can I modify data with this server?

No, this is read-only. You can query and inspect OpenVibeCoding data but cannot create, update, or delete runs, workflows, or other resources.

How do I install it?

Install via PyPI: `pip install openvibecoding-orchestrator`. Then configure it as an MCP server in your Claude or Cursor settings.

What authentication is required?

The README does not specify authentication requirements. Check the OpenVibeCoding documentation or installation guide for credential setup.

What is OpenVibeCoding?

OpenVibeCoding (Codeflow) is an orchestrator for long-running AI coding tasks. You provide a spec, it plans and executes the work, and you get back a branch, diff, and proof of work.

Is this free?

OpenVibeCoding is open-source under the MIT license, so yes—free and open to modify.

README (reference)

Source of truth, from the repository.

<p align="center"> <img src="https://em-content.zobj.net/source/apple/391/headphone_1f3a7.png" width="120" alt="headphone" /> </p> <h1 align="center">Codeflow</h1> <p align="center"> <strong>AI codes overnight, you ship in the morning</strong> </p> <p align="center"> <a href="https://github.com/xiaojiou176-open/codeflow/stargazers"><img src="https://img.shields.io/github/stars/xiaojiou176-open/codeflow?style=flat&color=yellow" alt="Stars"></a> <a href="https://github.com/xiaojiou176-open/codeflow/commits/main"><img src="https://img.shields.io/github/last-commit/xiaojiou176-open/codeflow?style=flat" alt="Last Commit"></a> <a href="LICENSE"><img src="https://img.shields.io/github/license/xiaojiou176-open/codeflow?style=flat" alt="License"></a> </p> <p align="center"> <a href="#what-you-get">What You Get</a> • <a href="#install">Install</a> • <a href="#how-it-work">How It Work</a> • <a href="#ecosystem">Ecosystem</a> </p>

Codeflow is the orchestrator for long-haul AI coding tasks. Hand it a spec, walk away, come back to a branch, a diff, and a proof of work.

┌──────────────────────────────────────┐
│  LOCAL-FIRST          ████████ 100%  │
│  SOURCE-TRACEABLE     ████████ 100%  │
│  TYPING REQUIRED      ░░░░░░░░   0%  │
│  VIBES                ████████ ZERO  │
│                                FILLER│
└──────────────────────────────────────┘

Long-running coding-agent orchestrator. Plan, run, replay, deliver.

What You Get

SurfaceWhat
orchestrator appsPlan it, run it, watch it, replay it.
packagesPluggable runners — Codex, Claude, Cline, OpenHands.
contracts & schemasSpec, plan, run, artifact. Every step versioned.
design systemCalm cockpit. No flashing red. No false confidence.
public skillsDrop into Claude/Codex/OpenClaw. Hand it a spec, get a branch.

[!IMPORTANT] Local-first by default. No silent telemetry. No cloud round-trip. Your data stays on your machine until you explicitly ship it somewhere.

Install

git clone https://github.com/xiaojiou176-open/codeflow.git
cd Codeflow
# follow the per-stack quickstart in INSTALL.md or docs/

Three commands. No curl | sh. No login. Read what you run.

Install break? Open your favorite agent and say "Read AGENTS.md and bootstrap Codeflow for me." Agent fix own brain. Long version: docs/.

How It Work

The repo is seven layers — exactly the seven commits in git log. New work goes in as small named PRs. No 50-file mystery commits.

LayerWhat
chore: scaffoldLicense, governance, hygiene gates, CI scaffolding.
feat(core)The primary engine. The reason Codeflow exists.
feat(modules)Packages, adapters, services, plugins. The second floor.
feat(contracts)Schemas, configs, public boundaries. Other code talks here.
test:Receipts. Everything in this layer must run.
feat(ops)Scripts, infra, CI helpers, build glue.
docs:Public docs surface. The pretty face.

git log reads like a building floor plan. Look once, know the whole shape.

Ecosystem

Codeflow lives in the open family: three open platforms. local-first, source-first, builder-first.

RepoWhat
OpenCampusCanvas + Gradescope + EdStem on one calm desk
Codeflow (you here)AI codes overnight, you ship in the morning
OpenUIStudiobrief in, React + shadcn out, proof attached

Cross-family taste: BeamMe · BrewMe · Codeflow · proofyard.

Star This Repo

If Codeflow saves you a click, an hour, or a headache — star costs zero. Fair trade. ⭐

Star History Chart

Also by Yifeng[Terry] Yu

  • OpenCampus — Canvas + Gradescope + EdStem on one calm desk
  • OpenUIStudio — brief in, React + shadcn out, proof attached
  • BeamMe — beam your agent config to any planet
  • BrewMe — wake up, news already brewed
  • proofyard — every claim ships with its receipt

License

MIT — small print, big freedom.

Related MCP servers

Compare-first local MCP sidecar for browser AI workflows.

2
TypeScript
MIT
View repository →

First-party Provenote MCP server for drafts, research threads, auditable runs, and knowledge search.

1
Python
MIT
View repository →
SOSourceHarbor MCP logo

Source-first AI knowledge control tower MCP server for YouTube, Bilibili, RSSHub, and RSS.

2
Python
MIT
View repository →

Local stdio MCP bridge for archive publishing, retrieval, and governance evidence.

1
Rust
MIT
View repository →

Spawn multiple specialized Claude Code agents as MCP servers, each with custom models, tools, and personalities.

16
TypeScript
ISC
View repository →

Spawn multiple specialized Codex agents as MCP servers, each with custom models, sandboxes, and personalities.

17
TypeScript
ISC
View repository →