coder-mcp-bridge
Active·★ 18·MIT·Updated 2026-08-15
Coder MCP Bridge is an event-driven multi-programming agent scheduling bridge designed for MCP clients like Codex. It unifies the control and observability of various backend agents (ZCode, OpenCode, Pi) under a consistent `agent-*` tool interface, while preserving their native session and inference capabilities.
#Multi-Agent Orchestration#Event-Driven#MCP Integration#Agent Scheduling#Resource Management#Session Management#LLM Agent Bridge
01
Features
01Unified Control Plane: Schedule ZCode, OpenCode, or Pi through the same MCP tools after one-time configuration.
02True Concurrency: Bridge does not limit global concurrency, allowing MCP schedulers to determine task splitting and parallelism.
03Resource Mutual Exclusion: Cross-process SQLite lease coordinates shared resources like worktrees, simulators, and build directories.
04Session Lifecycle Management: Supports recovery, guidance, interruption, branching, compaction, and explicit shutdown, with capabilities reported by each backend.
05Operational Observability: Unified exposure of reasoning, tool events, tokens, context, and final states.
02
Compatibility
Python
Runtime
Verified via docs
MCP Clients (e.g., Codex)
Integration
Verified via docs
ZCode
Backend Agent
Verified via docs
OpenCode
Backend Agent
Verified via docs
Pi
Backend Agent
Verified via docs
Docker
Deployment
Verified via docs
03
Quick start
1
$ git clone https://github.com/Deslord319/coder-mcp-bridge.git
2
$ cd coder-mcp-bridge
3
$ python3 server.py --probe
04
Use cases
↳Orchestrating multiple AI agents (ZCode, OpenCode, Pi) from a single MCP client for diverse tasks.
↳Managing concurrent agent tasks efficiently while ensuring proper serialization of conflicting resources.
↳Developing long-running, specification-driven agents with robust execution control and observable state transitions.
↳Benchmarking and comparing the performance, cost, and output quality of different AI agents for specific development tasks.
05
Alternatives
open-multi-agent-kit★ 148
Provider-neutral multi-agent control plane for coding agents. Route runtimes, scope MCP tools, run DAG workers, verify evidence, and replay agent runs from the omk CLI.
Agentflow★ 27
Production-grade framework for building multi-agent AI systems. Graph-based orchestration, LLM-agnostic (OpenAI, Google GenAI, Anthropic), 3-layer memory (Redis cache + Postgres + vector store), live agents, parallel tool execution, and native MCP. Ships a full ecosystem: backend, REST API + CLI, TypeScript SDK, and React playground
WRAI.TH★ 32
Multi-agent orchestration via MCP. Persistent memory, inter-agent messaging, goal cascade, context budget pruning. One binary, zero config.
cmuxlayer★ 30
cmux MCP server — 25 tools for terminal agent lifecycle management. Spawn, monitor, and orchestrate Claude, Codex, Cursor, Gemini, Kiro agents. Thinking-state detection, chunked input, layout management.
solace-agent-mesh★ 4.9k
An event-driven framework designed to build and orchestrate multi-agent AI systems. It enables seamless integration of AI agents with real-world data sources and systems, facilitating complex, multi-step workflows.
Comments
Log in to leave a comment
No comments yet. Be the first!