research-task-executor

What problem does it solve?

It solves the problem of reliably executing user-specified academic coding tasks by enforcing a structured, research-grade workflow with consistency checks and careful assumptions.

Core Features & Use Cases

• Precise trigger control: Only activates when the user inputs the exact phrase 科研启动-执行, preventing accidental or premature execution.
• Multi-agent implementation planning: Uses multiple expert agents to decompose, design, implement, and verify tasks without skipping steps.
• Research-code consistency guardrails: Performs environment checks and validates training/inference configuration alignment (including model naming and internal field verification).
• Project-aware execution: Reads PROGRESS.md, PIPELINE.md, EXAMPLES.csv, and ISSUES.md (when present) and manages git workflow and documentation updates after completion.

Quick Start

Say: 科研启动-执行 then describe the科研代码任务 you want completed.