omp-worker-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@omp-worker-mcpDelegate a DAG workflow to extract and summarize all API endpoints in the codebase"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
omp-worker-mcp
Durable Model Context Protocol (MCP) server for delegating asynchronous coding tasks and DAG workflows to local Oh My Pi (OMP) CLI sub-agents.
Quick Start • Configuration • Available Tools • Safety Contract • Upstream Attribution
Key Highlights
⚡ Asynchronous Delegated Execution: Offload heavy coding, refactoring, and exploration tasks to background OMP worker instances without blocking your main conversation session.
🔀 Topological DAG Orchestration: Execute interdependent batch tasks with automatic topological sorting, concurrency control, and dependency propagation.
🛡️ Workspace Path Isolation: Enforce explicit write-path boundaries and prevent overlapping file modifications between concurrent tasks.
🔍 Supervised Resumption & Envelopes: Inspect interim logs in real time, extract structured JSON outcome envelopes, and supply supervisory guidance to retry or adjust tasks.
Upstream Attribution & Disclaimer
External Upstream CLI: This project interfaces with Oh My Pi (OMP), an open-source tool released under the MIT License.
User Prerequisites:
Node.js >= 22.0.0.
Users must install and configure their own local instance of the OMP CLI.
Users are responsible for complying with the OMP CLI license and terms applicable to their environment.
Features
Asynchronous Delegated Execution: Launch independent sub-agent coding tasks without blocking the main conversational session.
DAG & Batch Orchestration: Run interdependent batch tasks with topological dependency sorting, concurrency limits, and automatic dependency propagation.
Strict Task Safety & Ownership: Built-in validation checks prevent multiple concurrent tasks from writing to overlapping file paths or colliding within the same workspace.
Continuous Supervision & Feedback: Inspect interim output, stream logs, retrieve structured JSON envelopes, and supply supervisory feedback to continue or retry attempts.
Cross-Platform Design: Designed for local Node.js environments on Windows, macOS, and Linux; validate OMP CLI availability on your platform.
Installation & Quick Start
# 1. Clone the repository
git clone https://github.com/divenire990/omp-worker-mcp.git
cd omp-worker-mcp
# 2. Install dependencies
npm ci
# 3. Build TypeScript to dist/
npm run build
# 4. Run test suite
npm testConfiguration
Configuration is managed via environment variables (or configured directly in your MCP client).
Variable | Description | Default |
| Path or executable name for the OMP CLI binary. |
|
| JSON array of arguments prepended to OMP CLI invocations. |
|
| Directory for storing job states, attempts, logs, and artifacts. |
|
| Optional custom browser automation instructions injected into task prompts. | (none) |
Cross-Platform Environment Examples
Windows (cmd / PowerShell)
set OMP_WORKER_OMP_COMMAND=omp
set OMP_WORKER_STATE_DIR=C:\Users\YourUser\.codex\state\omp-workermacOS / Linux (bash / zsh)
export OMP_WORKER_OMP_COMMAND=/usr/local/bin/omp
export OMP_WORKER_STATE_DIR=/home/youruser/.codex/state/omp-workerCodex MCP Client Configuration Example
Add the server to your Codex MCP configuration (e.g. in config.toml):
Windows Example
[mcp_servers.omp-worker]
command = "node"
args = ["C:/path/to/omp-worker-mcp/dist/index.js"]
[mcp_servers.omp-worker.env]
OMP_WORKER_OMP_COMMAND = "omp"
OMP_WORKER_STATE_DIR = "C:/Users/YourUser/.codex/state/omp-worker"macOS / Linux Example
[mcp_servers.omp-worker]
command = "node"
args = ["/path/to/omp-worker-mcp/dist/index.js"]
[mcp_servers.omp-worker.env]
OMP_WORKER_OMP_COMMAND = "/usr/local/bin/omp"
OMP_WORKER_STATE_DIR = "/home/youruser/.codex/state/omp-worker"Available MCP Tools
Tool | Purpose |
| Convenience tool: delegates a single task and waits up to |
| Spawns an asynchronous background worker for a coding task and immediately returns a |
| Waits for a running background job to finish or poll until a timeout is reached. |
| Fetches full attempt history, logs, outputs, and parsed structured envelopes for a job. |
| Feeds supervisory correction/guidance into a failed or blocked job for a new attempt. |
| Gracefully terminates a running job and its child process tree. |
| Spawns a dependency graph (DAG) of parallel/sequential batch tasks and waits for completion. |
| Waits for an asynchronous batch task group to progress or complete. |
| Cancels all running and queued tasks within a batch group. |
Task Safety & Ownership Contract
Write vs. Read-Only Isolation:
writetasks must declare the specific file paths they own.read_onlytasks are strictly restricted from workspace modification.
DAG Overlap Verification:
Tasks running in parallel cannot declare overlapping write boundaries.
Tasks modifying shared paths must declare explicit linear DAG dependencies (
depends_on).
Structured Verification Contract:
Every completed sub-task is expected to return structured verification details and updated artifact descriptions.
License
This project is licensed under the MIT License.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Agent-native collaboration network: orchestrate a team of long-running agents from any MCP client.
Durable agent-to-agent handoffs and shared scratchpad for multi-agent workflows.
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/divenire990/omp-worker-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server