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., "@z-code-mcp Orchestrate implementing user authentication with code analysis, code generation, and tests"
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.
MCP (Model Context Protocol) server for orchestrating multiple Claude Code sessions with HEAD/SUB coordination pattern.
Z-Code MCP enables complex, multi-task coding projects by coordinating multiple Claude Code CLI sessions:
HEAD Phase: Analyzes objectives and creates execution plans
SUB Phase: Executes tasks in parallel with controlled concurrency
Aggregation Phase: Combines results and generates summaries
Related MCP server: Orchestrator MCP Server
npx z-code-mcpFor always getting the latest version:
npx z-code-mcp@latestnpm install -g z-code-mcpAdd to your Claude configuration file:
Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"z-code-mcp": {
"command": "npx",
"args": ["-y", "z-code-mcp@latest"]
}
}
}For testing local changes without publishing:
{
"mcpServers": {
"z-code-mcp-dev": {
"command": "node",
"args": ["C:/path/to/z-code-mcp/build/index.js"]
}
}
}Variable | Description | Default |
| Claude CLI command |
|
| Max concurrent processes (1-10) |
|
| Temp directory for outputs | System temp |
| Keep temp files on error |
|
| Custom templates directory | Built-in |
| Log level (debug/info/warn/error) |
|
Template | Description |
| Analyze existing code structure and patterns |
| Generate new code based on specifications |
| Write tests for existing code |
| Create documentation |
| Security analysis and vulnerability detection |
| Code improvements and refactoring |
| Use custom template |
Once configured, the MCP exposes an orchestrate tool:
{
"objective": "Implement user authentication feature",
"workingDirectory": "/path/to/project",
"tasks": [
{
"id": "analyze",
"type": "code_analysis",
"description": "Analyze current auth implementation"
},
{
"id": "implement",
"type": "code_generation",
"description": "Implement JWT authentication",
"dependencies": ["analyze"]
},
{
"id": "test",
"type": "test_writing",
"description": "Write auth tests",
"dependencies": ["implement"]
}
],
"coordination": {
"strategy": "dependency-based",
"maxConcurrent": 3
}
}# Install dependencies
npm install
# Build
npm run build
# Watch mode
npm run dev
# Run MCP Inspector
npm run inspectornpx caches packages: Use npx z-code-mcp@latest to force latest version
Local development: Use direct path reference for immediate updates after build
MIT
Orchestrate multiple Claude Code sessions with HEAD/SUB coordination. HEAD analyzes the objective, creates an execution plan, spawns SUB processes to handle individual tasks, and aggregates results.
| Name | Required | Description | Default |
|---|---|---|---|
| tasks | Yes | List of tasks to execute | |
| objective | Yes | High-level objective for the orchestration | |
| coordination | No | Coordination settings | |
| resourceLimits | No | Resource limits | |
| workingDirectory | Yes | Base directory for all operations |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden for behavioral disclosure. While it mentions the HEAD/SUB pattern and task handling, it does not explain important aspects such as error handling, recovery strategies, authentication requirements, resource cleanup, or side effects. The description provides a high-level overview without sufficient depth for an agent to anticipate behavior beyond the basic workflow.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with two well-structured sentences. The first sentence introduces the core functionality and the HEAD/SUB concept, while the second elaborates on the workflow. Every sentence contributes meaningfully, and there is no redundancy or unnecessary information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of the tool (nested parameters, coordination strategies, resource limits) and the absence of an output schema, the description is too high-level. It does not cover return value structure, failure modes, aggregation details, or how task dependencies are handled. The agent lacks sufficient context to understand the full orchestration lifecycle, which is a significant gap for such a nuanced tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description adds minimal context beyond the schema; it does not explain how the parameters (e.g., tasks, coordination strategy) affect the HEAD/SUB orchestration or the execution plan. The description's value beyond the schema is limited, but the schema itself is thorough.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the tool orchestrates Claude Code sessions with HEAD/SUB coordination, explaining the high-level workflow of analyzing objectives, creating execution plans, spawning SUB processes, and aggregating results. The verb 'orchestrate' and resource 'multiple Claude Code sessions' are specific and unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit when-to-use or when-not-to-use guidance is provided. The description implies usage for tasks requiring multi-session coordination but does not specify prerequisites, alternatives, or conditions for deferring. Since there are no sibling tools, the lack of explicit guidelines is less critical but still a missed opportunity.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v1.0.1orchestrateWith only one tool, there is no possibility of confusion between tools. The tool's purpose is clearly defined.
Single tool named with a verb ('orchestrate'), consistent with a command-style action. No pattern inconsistency possible.
A single tool for a server that coordinates multiple sessions is borderline thin. While the tool itself is non-trivial, the server would benefit from additional tools for session management, making the count slightly low for its apparent scope.
The tool set is severely incomplete. The server's purpose implies managing multiple Claude Code sessions, but there are no tools for creating, listing, monitoring, or terminating sessions—only a single orchestration tool that attempts to handle everything.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
AI work orchestration for plans, tasks, teams, and coding-agent dispatch.
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
The team layer for AI coding agents: shared contracts, collision alerts, E2EE sessions.
Task tracking built for coding agents. Work is leased, so two agents never take the same SubTask.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Zeliper/z-code-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
