blaze

分类: 数据与AI | 上传者: srpadrono | 下载: 0 | 版本: v1.0（最新）

Generates Mermaid flowcharts and decision trees from journeys.json showing test coverage with before/after comparison. Use when the user says /blaze or wants to visualize coverage.

更新日志: Source: GitHub https://github.com/srpadrono/Pathfinder

目录结构

当前层级: skills/blaze/

📄 SKILL.md 4.2 KB

SKILL.md

---
name: blaze
description: "Generates Mermaid flowcharts and decision trees from journeys.json showing test coverage with before/after comparison. Use when the user says /blaze or wants to visualize coverage."
---

# Blazing

Transform the journey map into visual Mermaid diagrams that show exactly what's tested and what's not.

## Contents
- [Process](#process)
- [Baseline Management](#baseline-management)
- [Updating After Tests](#updating-after-tests)
- [Mermaid Syntax Notes](#mermaid-syntax-notes)
- [Error Handling](#error-handling)

## Process

1. **Read journey map:** Load `<testDir>/pathfinder/journeys.json`
2. **Generate diagrams:** Run `python3 "${CLAUDE_SKILL_DIR}/../pathfinder/scripts/generate-diagrams.py" <testDir>/pathfinder/journeys.json`

This creates `<testDir>/pathfinder/blazes.md` with multiple diagram types:

### Legend
A table at the top explains all symbols, colours, and arrow types used:

| Symbol | Meaning |
|--------|---------|
| 🟢 ✅ | **Tested** — step has a passing UI test |
| 🟡 ⚠️ | **Partial** — test written but disabled or implicitly covered |
| 🔴 ❌ | **Untested** — no UI test coverage |
| 🔵 🔀 | **Decision point** — user chooses between paths |
| ⚡ | **Error path** — API failure branch |
| `──▶` | Same-screen transition |
| `╌╌▶` | Cross-screen navigation |

### Combined Decision Tree
All journeys merged into one flowchart showing every branching path:
- Diamond `{{"🔀"}}` nodes at decision points where the user chooses
- Shared prefixes merged (e.g., if two journeys start at the same screen)
- Error paths branch from loading/API steps with dashed `⚡ API Error` arrows
- Spot coverage gaps at a glance in the full tree

### Per-Journey Flowcharts
One Mermaid `flowchart TD` per journey with colour-coded nodes.

**Node shapes by status:**
| Status | Shape | Colour |
|--------|-------|--------|
| Tested | `("label")` rounded | 🟢 `#2ea043` green |
| Partial | `[/"label"/]` parallelogram | 🟡 `#d29922` amber |
| Untested | `["label"]` rectangle | 🔴 `#f85149` red |
| Decision | `{{"label"}}` diamond | 🔵 `#1f6feb` blue |

**Arrow types:**
| Type | When used |
|------|-----------|
| `-->` solid | Steps on the same screen |
| `-.->` dashed | Navigation across screens |
| `-.->|"⚡ API Error"|` labelled dashed | Error branch from API/loading step |

### Before/After Comparison (automatic)
On first run, a baseline is auto-saved to `<testDir>/pathfinder/journeys-baseline.json`.
On subsequent runs, if coverage has changed, the output includes:

- **Before (Baseline)** — decision tree at baseline coverage
- **After (Current)** — decision tree with current coverage
- **Coverage Delta** — summary table with per-journey step deltas

## Baseline Management

| Command | What it does |
|---------|-------------|
| `python3 "${CLAUDE_SKILL_DIR}/../pathfinder/scripts/generate-diagrams.py" journeys.json` | Normal run; auto-creates baseline on first run |
| `python3 "${CLAUDE_SKILL_DIR}/../pathfinder/scripts/generate-diagrams.py" journeys.json --save-baseline` | Reset baseline to current coverage |
| `python3 "${CLAUDE_SKILL_DIR}/../pathfinder/scripts/generate-diagrams.py" journeys.json --clear-baseline` | Remove baseline; next run creates a fresh one |

Use `--save-baseline` after completing a coverage sprint to reset the "before" snapshot for the next round.

3. **Commit:** `git add <testDir>/pathfinder/blazes.md && git commit -m "Diagram: N journeys mapped (X% coverage)"`

## Updating After Tests

After writing tests in the scout phase, re-run:
```bash
python3 "${CLAUDE_SKILL_DIR}/../pathfinder/scripts/generate-diagrams.py" <testDir>/pathfinder/journeys.json
```

The diagram updates for newly tested steps, and the before/after comparison shows progress.

## Mermaid Syntax Notes

- Parentheses `()` in labels are auto-escaped to `[]` to avoid Mermaid shape conflicts
- All labels are wrapped in double quotes for safety
- If you encounter parse errors, check for unescaped special characters in step action text

## Error Handling

- If `journeys.json` is missing, invoke `/map` first.
- If a journey has 0 steps, it was mapped incorrectly — go back and re-examine that flow.
- If Mermaid parse errors occur, re-run the generator — it sanitizes labels automatically.

登录后下载/点赞/收藏 ❤ 9 | ★ 0

blaze

目录结构

SKILL.md

举报内容

提示