Home › Other › MCP Task Orchestrator

MCP Task Orchestrator

Server-enforced workflow discipline for AI agents: work items, dependency graphs, quality gates

Kotlin

★ 206 stars Last pushed 2026-09-24 License: MIT

View repository →

Connect

Review any command before running it. Package names and URLs come from the server's own registry entry.

Package (oci)

docker run -i --rm ghcr.io/jpicklyk/task-orchestrator:3.14.0

Or add to your MCP client config:

{
  "mcpServers": {
    "task-orchestrator": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/jpicklyk/task-orchestrator:3.14.0"
      ],
      "env": {
        "DATABASE_PATH": "<YOUR_VALUE>",
        "USE_FLYWAY": "<YOUR_VALUE>",
        "AGENT_CONFIG_DIR": "<YOUR_VALUE>",
        "MCP_TRANSPORT": "<YOUR_VALUE>",
        "MCP_HTTP_PORT": "<YOUR_VALUE>",
        "LOG_LEVEL": "<YOUR_VALUE>"
      }
    }
  }
}

DATABASE_PATH — SQLite database file path (default: data/current-tasks.db)
USE_FLYWAY — Enable Flyway database migrations (default: true)
AGENT_CONFIG_DIR — Directory containing .taskorchestrator/config.yaml for workflow schemas (default: working dir)
MCP_TRANSPORT — Transport mode: stdio (default) or http
MCP_HTTP_PORT — HTTP port when using http transport (default: 3001)
LOG_LEVEL — Logging level: DEBUG, INFO, WARN, ERROR (default: INFO)