メニュー
Display PRD files in human-readable format for review and status tracking
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Research and recommend the best local LLM model stack for the current machine. Detects hardware (CPU, GPU, RAM, VRAM), discovers installed hosting stacks (Ollama, llama.cpp, vLLM, etc.), searches Hugging Face for compatible models, and enables easy downloads. Use when the user wants to find, evaluate, or download local AI models that will run well on their system.
Zellij terminal multiplexer — creating KDL layouts, managing sessions via CLI, and running commands without disrupting the user's workspace
Use when adding a new service to a Nix flake — creating packages from upstream repos (Go, Node, Python), writing service modules, managing configs, wiring secrets, and testing. Covers the full lifecycle from source to running daemon.
Use when a launchd service on nix-darwin fails to start, exits with non-zero status, or doesn't restart after a switch. Triggers on: exit code 78, "EX_CONFIG", "Operation not permitted", daemon vs user.agent decisions, plist not being reloaded.
Use when managing 1Password secrets via Nix on nix-darwin. Triggers on: adding API keys, tokens, or credentials to service configs; wiring opnix 1Password items into home-manager modules; creating opnix-managed secret files with activation script ordering.
Use when tracking, planning, implementing, or reviewing work using yx (yaks) with the autonomous /shave loop, or when multiple agents need to coordinate on shared tasks
| name | prd-review |
| description | Display PRD files in human-readable format for review and status tracking |
Convert prd.json files into human-readable format for review, status tracking, and progress monitoring. Helps humans understand and validate PRDs before and during Ralph execution.
Display complete PRD with all details:
Review the PRD at prd.json in human-readable format
Quick overview of completion status:
Show PRD status summary for prd.json
Deep dive into one story:
Show details for story US-003 from prd.json
When asked to review a PRD, output in this format:
# PRD: [project]
**Branch:** `[branchName]`
**Description:** [description]
## Progress
[■■■■■□□□□□] 5/10 stories complete (50%)
## Stories
### ✅ US-001: [title] (Priority 1)
**Status:** COMPLETE
> [description]
**Acceptance Criteria:**
- ✅ [criterion 1]
- ✅ [criterion 2]
**Notes:** [any notes from agent]
---
### 🔄 US-002: [title] (Priority 2)
**Status:** IN PROGRESS (current)
> [description]
**Acceptance Criteria:**
- ✅ [completed criterion]
- ⬜ [pending criterion]
---
### ⬜ US-003: [title] (Priority 3)
**Status:** PENDING
> [description]
**Acceptance Criteria:**
- ⬜ [criterion 1]
- ⬜ [criterion 2]
# [project] - Status Summary
**Branch:** `[branchName]`
**Progress:** [■■■■■□□□□□] 5/10 (50%)
| ID | Title | Priority | Status |
|----|-------|----------|--------|
| US-001 | [title] | 1 | ✅ Complete |
| US-002 | [title] | 2 | 🔄 In Progress |
| US-003 | [title] | 3 | ⬜ Pending |
**Next Up:** US-003 - [title]
**Estimated Remaining:** 5 stories
# US-003: [title]
**Priority:** 3 of 10
**Status:** ⬜ Pending
**Dependencies:** US-001, US-002
## Description
> As a [role], I want [goal] so that [benefit].
## Acceptance Criteria
| # | Criterion | Status |
|---|-----------|--------|
| 1 | [criterion text] | ⬜ |
| 2 | [criterion text] | ⬜ |
| 3 | Typecheck passes | ⬜ |
## Size Analysis
- **Files affected:** ~2-3 (estimated)
- **Complexity:** Medium
- **Est. iterations:** 1-2
## Notes
[Any notes from previous iterations]
Use these consistently:
| Symbol | Meaning |
|---|---|
| ✅ | Complete (passes: true) |
| 🔄 | In Progress (first passes: false by priority) |
| ⬜ | Pending (not yet started) |
| ❌ | Blocked (has failing dependency) |
Generate progress bar based on completion:
0%: [□□□□□□□□□□]
10%: [■□□□□□□□□□]
50%: [■■■■■□□□□□]
100%: [■■■■■■■■■■]
When reviewing, check for issues:
Flag if a story depends on incomplete higher-priority stories:
⚠️ **Dependency Warning**
US-005 (Priority 5) may depend on:
- US-003 (Priority 3) - ⬜ Not complete
- US-004 (Priority 4) - ⬜ Not complete
Consider completing dependencies first.
Flag stories that nothing depends on (potential scope creep):
ℹ️ **Note:** US-008 appears to be standalone (no other stories depend on it).
This is fine if intentional, but verify it's in scope.
When reviewing, flag potential issues:
⚠️ **Size Warning:** US-004 has 7 acceptance criteria.
Consider splitting into smaller stories.
⚠️ **Vague Criterion:** "Code is well-organized" in US-003
This is not machine-verifiable. Suggest specific check.
⚠️ **Missing Verification:** US-005 has no typecheck/test criterion.
Add "Typecheck passes" or equivalent.
{
"project": "TaskApp",
"branchName": "ralph/priorities",
"description": "Add priority system",
"userStories": [
{"id": "US-001", "title": "Add priority column", "passes": true, ...},
{"id": "US-002", "title": "Add API endpoint", "passes": false, ...},
{"id": "US-003", "title": "Add UI component", "passes": false, ...}
]
}
# PRD: TaskApp
**Branch:** `ralph/priorities`
**Description:** Add priority system
## Progress
[■■■□□□□□□□] 1/3 stories complete (33%)
## Stories
### ✅ US-001: Add priority column (Priority 1)
**Status:** COMPLETE
...
### 🔄 US-002: Add API endpoint (Priority 2)
**Status:** IN PROGRESS
...
### ⬜ US-003: Add UI component (Priority 3)
**Status:** PENDING
...
Review prd.json and flag any issues before I run Ralph
Show current Ralph progress for prd.json
Summarize what Ralph completed in prd.json
ralph-specs - Write Ralph-compatible specificationsverification-before-completion - Run verification commands