Skip to content
Back to skills

Oh My Ccg Research

ASecurity

RPI 研究阶段 — 需求转化为约束集

  • 666 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 2, 2026
ai-agentsbashfrontendbackend

Works with

  • cli
  • mcp

Security analysis

A100/100

Scanned September 2, 2026

npx -y skills add majiayu000/claude-skill-registry --skill oh-my-ccg-research --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Oh My Ccg Research?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Oh My Ccg Research
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majiayu000-oh-my-ccg-research/badge)](https://www.skillsdirectory.com/skills/majiayu000-oh-my-ccg-research)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: oh-my-ccg-research
description: RPI 研究阶段 — 需求转化为约束集
---

# oh-my-ccg Research Phase

You are executing the RPI Research phase. Transform requirements into verifiable constraint sets.

## Prerequisites
- RPI state must exist (run `/oh-my-ccg:init` first)
- A requirement description must be provided

## Workflow

### Step 1: Restore State
Use the **oh-my-ccg-tools** MCP server's `rpi_state_read` tool.
Confirm current phase allows transition to RESEARCH.

### Step 2: Create OpenSpec Change (Optional)
If OpenSpec CLI is available:
```bash
npx openspec new change "<requirement-name>"
```
If OpenSpec is not installed, skip this step — artifacts will be managed via file system in `.oh-my-ccg/plans/`.

### Step 3: Codebase Exploration
Use the `explore` agent to scan the codebase for:
- Existing patterns related to the requirement
- Files that will be affected
- Dependencies and interfaces involved

### Step 4: Multi-Model Constraint Exploration (PARALLEL)
Launch both models **simultaneously in a single message** using MCP tools:

**Codex** (backend perspective) — use `ask_codex` tool:
- `agent_role`: "analyst"
- `prompt`: Describe the requirement and ask for backend/logic constraints
- `context_files`: Relevant source files from Step 3
- `background`: true

**Gemini** (frontend perspective) — use `ask_gemini` tool:
- `agent_role`: "designer"
- `prompt`: Describe the requirement and ask for UI/UX constraints
- `files`: Relevant source files from Step 3
- `background`: true

Then use `check_job_status` for both job IDs to collect results.

### Step 5: Synthesize Constraints
Merge findings from explore + Codex + Gemini into a unified constraint set:
- Classify each as Hard or Soft
- Define verification criteria
- Resolve any conflicts between models

### Step 6: User Interaction
Present constraints to user. Ask for:
- Confirmation of hard constraints
- Prioritization of soft constraints
- Clarification of any ambiguities

### Step 7: Persist State
Use the **oh-my-ccg-tools** MCP server's `rpi_state_write` tool to save:
- Updated constraints array
- Phase transition to "research"
- Change ID (if OpenSpec was used)

If OpenSpec is available, also generate `proposal.md` in the change directory.

### Step 8: Report
Output summary: constraint count (Hard/Soft), key findings, and instructions to `/clear` then `/oh-my-ccg:plan`.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.