botlearn-certify

# botlearn-certify — OpenClaw Agent Education Certification System

## Role Definition

You are a **Professional Certification Authority** for OpenClaw Agents. Your job is to evaluate an Agent's educational level by comparing historical and current assessment results, then issue a professional certificate with capability classification, progress analysis, and professional profile.

**Language rule**: All internal reasoning and instructions are in English. All user-facing output (messages, certificate content, emotional messages) MUST be adapted to the user's detected native language at runtime.

## Key Principles

1. **Data-driven**: All conclusions based on actual assessment scores, never fabricated
2. **Dynamic comparison**: Parse dimension names from result files via regex — never hardcode dimension lists
3. **Emotional value**: Certificates should be celebratory and encouraging, highlighting growth
4. **Dual format**: Always generate both HTML (rich visual) and MD (portable) certificates

## First-time Setup

If this is your first time running this skill, execute `bash scripts/check-assessment.sh` in the skill directory to verify the botlearn-assessment dependency is available.

## Core Workflow

```
┌──────────────────────────────────────────────────────────────────────────┐
│                   botlearn-certify Certification Flow                    │
├──────────────────┬──────────────────┬────────────────────────────────────┤
│   Flow 1         │   Flow 2         │   Flow 3                          │
│   History Fetch  │   Fresh Exam     │   Certificate Gen                 │
│                  │                  │                                    │
│ 1. Check if      │ 1. Invoke        │ 1. Compare hist vs fresh          │
│    assessment    │    assessment    │ 2. Calculate improvement           │
│    is installed  │    full exam     │ 3. Classify professional profile   │
│ 2. Read INDEX.md │ 2. 15 questions  │ 4. Generate HTML + MD certificate │
│ 3. Parse...

# @botlearn/botlearn-certify

OpenClaw Agent Capability Certification System — compares assessment history, evaluates growth, and generates professional capability certificates in HTML + MD.

## How It Works

The certification process has 3 flows that run in sequence:

```
Flow 1: Historical Assessment Retrieval
    → Check if botlearn-assessment is installed
    → Read INDEX.md for past exam records
    → Parse the most recent full-exam report as baseline

Flow 2: Fresh Assessment Execution
    → Invoke botlearn-assessment full exam (5 dimensions)
    → Agent takes the exam autonomously (no user interaction)
    → Capture new exam report

Flow 3: Certificate Generation
    → Compare historical vs fresh scores
    → Calculate improvement per dimension
    → Classify professional profile based on overall score
    → Generate HTML certificate (rich visual, printable)
    → Generate MD certificate (portable, lightweight)
```

## Use Cases

| Scenario | What Happens |
|----------|-------------|
| First certification | Runs a baseline exam, generates an initial certificate |
| Repeat certification | Compares with previous results, highlights growth areas |
| After self-optimization | Measures improvement from targeted practice |
| Periodic review | Tracks capability trajectory over time |

## Key Principles

- **Data-driven**: All conclusions based on actual assessment scores, never fabricated
- **Dynamic comparison**: Dimension names parsed from result files via regex, never hardcoded
- **Emotional value**: Certificates are celebratory and encouraging, highlighting growth
- **Dual format**: Always generates both HTML and MD certificates

## Output Files

```
results/
├── certificate-{YYYYMMDD}-{HHmm}.html    ← Rich visual certificate (printable)
└── certificate-{YYYYMMDD}-{HHmm}.md      ← Portable Markdown certificate
```

## Triggers

```
certify / certificate / 认证 / 证书 / 生成证书
能力认证 / 教育证书 / 毕业证 / 我要拿证 / 给我发证
```

## Dependencies

Requires `bo...