An MCP server that manages features and tasks across multiple projects. Useful for having the AI take PRD documents, breaking them into features and tasks for use across multiple interactions.
A Kotlin implementation of the Model Context Protocol (MCP) server for comprehensive task management, providing AI assistants with a structured, context-efficient way to interact with project data.
# Pull latest release
docker pull ghcr.io/jpicklyk/task-orchestrator:latest
# Or specific version
docker pull ghcr.io/jpicklyk/task-orchestrator:1.0.1
# Or latest build from main branch
docker pull ghcr.io/jpicklyk/task-orchestrator:main
# Build locally
./scripts/docker-clean-and-build.bat # Windows
# Or manually: docker build -t mcp-task-orchestrator:dev .
Add to your claude_desktop_config.json
:
Production Configuration
{
"mcpServers": {
"task-orchestrator": {
"command": "docker",
"args": [
"run", "--rm", "-i",
"--volume", "mcp-task-data:/app/data",
"ghcr.io/jpicklyk/task-orchestrator:latest"
]
}
}
}
Local Development Configuration
{
"mcpServers": {
"task-orchestrator": {
"command": "docker",
"args": [
"run", "--rm", "-i",
"--volume", "mcp-task-data:/app/data",
"mcp-task-orchestrator:dev"
]
}
}
}
Use the JSON configuration command:
# Production version (latest release)
claude mcp add-json task-orchestrator '{"type":"stdio","command":"docker","args":["run","--rm","-i","-v","mcp-task-data:/app/data","ghcr.io/jpicklyk/task-orchestrator:latest"]}'
# Specific version
claude mcp add-json task-orchestrator '{"type":"stdio","command":"docker","args":["run","--rm","-i","-v","mcp-task-data:/app/data","ghcr.io/jpicklyk/task-orchestrator:1.0.1"]}'
# Latest from main branch
claude mcp add-json task-orchestrator '{"type":"stdio","command":"docker","args":["run","--rm","-i","-v","mcp-task-data:/app/data","ghcr.io/jpicklyk/task-orchestrator:main"]}'
# Local development version (after building locally)
claude mcp add-json task-orchestrator '{"type":"stdio","command":"docker","args":["run","--rm","-i","-v","mcp-task-data:/app/data","mcp-task-orchestrator:dev"]}'
# Test the Docker container runs correctly
docker run --rm -i -v mcp-task-data:/app/data ghcr.io/jpicklyk/task-orchestrator:latest
# Test MCP connection (requires Node.js)
node scripts/test-mcp-connection.js
Ask Claude:
Project (optional)
โโโ Feature (optional)
โโโ Task (required) โโ Dependencies โ Task
โโโ Section (optional, detailed content)
create_feature_workflow
- Comprehensive feature creationtask_breakdown_workflow
- Complex task decompositionbug_triage_workflow
- Systematic bug managementproject_setup_workflow
- Complete project initializationimplement_feature_workflow
- Git-aware feature implementation with completion validation# Build
./gradlew build
# Run
java -jar build/libs/mcp-task-orchestrator-*.jar
# Configure environment for local development
MCP_TRANSPORT=stdio
DATABASE_PATH=data/tasks.db
USE_FLYWAY=true
MCP_DEBUG=true # Enable debug logging
Variable | Description | Default |
---|---|---|
MCP_TRANSPORT | Transport type | stdio |
DATABASE_PATH | SQLite database path | data/tasks.db |
USE_FLYWAY | Enable Flyway database migrations | true |
MCP_SERVER_NAME | Server name | mcp-task-orchestrator |
MCP_DEBUG | Enable debug logging | false |
Version follows semantic versioning with git-based build numbers:
{major}.{minor}.{patch}.{git-commit-count}-{qualifier}
1.0.0.123
)1.0.0.123-beta-01
)Current versioning defined in build.gradle.kts.
# Run tests
./gradlew test
# Test MCP connection
node scripts/test-mcp-connection.js
# Debug mode
MCP_DEBUG=true java -jar build/libs/mcp-task-orchestrator-*.jar
MCP_DEBUG=true
and check logs in logs/
docker volume inspect mcp-task-data
See contributing guidelines for details.
{ "mcpServers": { "task-orchestrator": { "command": "docker", "args": [ "run", "--rm", "-i", "--volume", "mcp-task-data:/app/data", "ghcr.io/jpicklyk/task-orchestrator:latest" ] } } }
Related projects feature coming soon
Will recommend related projects based on sub-categories