Skip to main content
Glama

mcp-chain-of-thought

index.md3.45 kB
**Please strictly follow the guidelines below** ## Solution Evaluation ### Task Summary {summary} ### Analysis Results {analysis} ## Evaluation Points ### 1. Technical Integrity - Check for technical flaws and logical loopholes in the solution - Validate edge case and exception handling - Confirm data flow and control flow integrity - Assess the reasonableness of technology choices ### 2. Performance and Scalability - Analyze resource usage efficiency and optimization potential - Evaluate system load scalability - Identify potential optimization points - Consider future functional expansion possibilities ### 3. Requirement Compliance - Verify the implementation status of functional requirements - Check compliance with non-functional requirements - Confirm the accuracy of requirement understanding - Evaluate user experience and business process integration ## Decision Points Choose subsequent actions based on the evaluation results: - **Found critical issues**: Use "analyze_task" to resubmit an improved solution - **Minor adjustments**: Apply these small improvements in the next execution step - **Solution sound**: Use "split_tasks" to decompose the solution into executable subtasks. If there are too many tasks or the content is too long, use the "split_tasks" tool multiple times, submitting only a small portion of tasks each time. ## split_tasks Update Mode Selection - **append** - Keep all existing tasks and add new ones - **overwrite** - Clear unfinished tasks, keep completed ones - **selective** - Selectively update specific tasks, keep others - **clearAllTasks** - Clear all tasks and create a backup ## Knowledge Transfer Mechanism 1. **Global Analysis Result** - Link the complete analysis document 2. **Task-Specific Implementation Guide** - Save specific implementation methods for each task 3. **Task-Specific Verification Criteria** - Set clear verification requirements ## Task Splitting Guide (Please strictly follow these rules) - **Atomicity**: Each subtask should be independently operable or testable - **Dependency**: If a task depends on others, mark the "dependencies" field - **Moderate Splitting**: Avoid over-granularity (too small) or over-consolidation (too large) - **Consolidate when necessary**: If the modifications are minor or not complex, integrate them appropriately with other tasks to avoid excessive tasks due to over-simplification - **Repeated Calls**: If too many tasks or long content prevents "split_tasks" from working correctly, use the tool multiple times, submitting only a small portion of tasks each time - **Simplify Tasks**: If adding only one task at a time still doesn't work, consider further splitting or simplifying the task while retaining core content - **JSON Format Note**: - **No Comments**: JSON format itself does not support comments (e.g., `#` or `//`). Adding comments within JSON content will lead to parsing failures. - **Escape Characters**: Special characters within JSON content, such as double quotes (`\"`) and backslashes (`\\\\`), must be properly escaped. Otherwise, they will be treated as illegal characters, causing parsing failures. **Serious Warning**: The parameters you pass each time you call split_tasks cannot exceed 5000 characters. If it exceeds 5000 characters, please call the tool multiple times to complete. **Now start calling the `split_tasks` or `analyze_task` tool** **Using tools is strictly required.**

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/liorfranko/mcp-chain-of-thought'

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