mirror of
https://github.com/GuDaStudio/commands.git
synced 2026-03-22 19:18:52 +08:00
2.1 KiB
2.1 KiB
name: GudaSpec: Implementation
description: Execute approved OpenSpec changes via multi-model collaboration with Codex/Gemini.
category: GudaSpec
tags: [openspec, implementation, multi-model, codex, gemini]
Guardrails
- If the project is detected to lack
./openspec/dir, prompt the user to initialize the project using/gudaspec:init. - Never apply external model prototypes directly—all Codex/Gemini outputs serve as reference only and must be rewritten into readable, maintainable, production-grade code.
- Keep changes tightly scoped to the requested outcome; enforce side-effect review before applying any modification.
- Minimize documentation—avoid unnecessary comments; prefer self-explanatory code.
Steps
- Run
openspec viewto inspect current project status and reviewActive Changes; UseAskUserQuestionstool ask the user to confirm which proposal ID they want to implement and wait for explicit confirmation before proceeding. - Run
/opsx:apply <proposal_id>and then follow it. - When performing a specific coding task, refer to the code prototype provided by an effective model:
- Route A: Gemini Kernel —
mcp__gemini__geminitool for frontend/UI/styling tasks (CSS, React, Vue, HTML, component design). - Route B: Codex Kernel —
mcp__codex__codextool for backend/logic/algorithm tasks (API, data processing, business logic, debugging). - Mandatory constraint: When communicating with Codex/Gemini, the prompt must explicitly require returning a
Unified Diff Patchonly; external models are strictly forbidden from making any real file modifications.
- Route A: Gemini Kernel —
- Upon receiving the diff patch from Codex/Gemini, NEVER apply it directly; rewrite the prototype by removing redundancy, ensuring clear naming and simple structure, aligning with project style, and eliminating unnecessary comments.
- For each completed task, conduct multi-model reviews using Codex
mcp__codex__codex/ Geminimcp__gemini__gemini, requiring iterative reviews until receiving dual-model LGTM approval. - MUST follow the
/opsx:apply <proposal_id>.