Skip to main content
Glama
index.md1.56 kB
## Task Splitting - {updateMode} Mode ## Splitting Strategy 1. **Functional Decomposition** - Independent testable sub-functions with clear inputs and outputs 2. **Technical Layer Decomposition** - Separate tasks along architectural layers, ensuring clear interfaces 3. **Development Stage Decomposition** - Core functionality first, optimization features later 4. **Risk-based Decomposition** - Isolate high-risk parts, reduce overall risk ## Task Quality Review 1. **Task Atomicity** - Each task is small and specific enough to be completed independently 2. **Dependencies** - Task dependencies form a directed acyclic graph, avoiding circular dependencies 3. **Description Completeness** - Each task description is clear and accurate, including necessary context ## Task List {tasksContent} ## Dependency Management - Dependencies can be set using task names or task IDs - Minimize the number of dependencies, only set direct prerequisite tasks - Avoid circular dependencies, ensure the task graph is directed and acyclic - Balance the critical path, optimize possibilities for parallel execution ## Decision Points - If task splitting is found unreasonable: call "split_tasks" again to adjust - If task splitting is confirmed to be sound: generate execution plan, determine priorities **Severe Warning** Each time you call split_tasks, the parameters you pass cannot exceed 5000 characters. If it exceeds 5000 characters, please call the tool multiple times to complete **If there are remaining tasks, please continue to call "split_tasks"**

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/cjo4m06/mcp-shrimp-task-manager'

If you have feedback or need assistance with the MCP directory API, please join our Discord server