Task Decomposer
Break complex tasks into structured subtasks with dependency analysis, parallel execution groups, and effort estimation.
Get the CLI Tool
Run the task decomposer locally as an MCP server, or try it online below.
npx @clinetools/task-decompose
- Automatic subtask identification from natural language descriptions
- Dependency inference — research, implementation, testing, deployment phases
- Parallel execution groups — find tasks that can run concurrently
- Critical path analysis — identify the bottleneck chain
- Effort estimation — small, medium, large per subtask
How to Use It
Three ways to decompose tasks — pick the one that fits your workflow.
Try Online
Use the interactive demo below to paste a task description and see it broken down instantly.
Use via CLI
Run as a local MCP server and connect any MCP-compatible client.
Add to Cline / Claude Code
Add the tool to your MCP settings for instant access from your AI assistant.
MCP Client Configuration (Cline)
{
"mcpServers": {
"task-decompose": {
"command": "npx",
"args": ["@clinetools/task-decompose"]
}
}
}
Claude Code Configuration
# In your project's .mcp.json:
{
"mcpServers": {
"task-decompose": {
"command": "npx",
"args": ["@clinetools/task-decompose"]
}
}
}
Example Prompt: Decompose a Feature
// Prompt to your AI agent:
"Break down this task: Add user authentication
with OAuth2, store profiles in the DB, create a
settings page, and write tests"
// The agent calls:
decompose_task({
task: "Add user authentication with OAuth2...",
maxDepth: 2,
includeEstimates: true
})
Try It Online
Describe a complex task and watch it decompose into structured subtasks with dependencies.
Describe Your Task
Enter a task description in natural language
Describe a task and click Decompose Task to see the breakdown.
Subtask Breakdown
Understanding Task Decomposition
Why breaking tasks down leads to better planning and faster delivery.
Task Decomposition Benefits
Large tasks feel overwhelming. Breaking them into small, concrete subtasks makes progress visible, reduces cognitive load, and reveals hidden complexity before you start coding.
Dependency Graphs
Understanding which tasks depend on others prevents wasted work and blocked developers. The decomposer automatically infers that testing depends on implementation, which depends on research.
Parallel Execution
Not all tasks need to be done sequentially. The decomposer identifies groups of subtasks that can run concurrently, helping teams and AI agents work on multiple fronts simultaneously.
Effort Estimation
Each subtask receives a small, medium, or large effort estimate based on complexity keywords and scope. The critical path shows the minimum total time needed, even with unlimited parallelism.
Decompose Every Task
Never start a complex task without a plan. Our Pro plan includes AI-enhanced decomposition with custom templates.
View Plans