Plugin Developer Agent

Implementation Expertise

I guide the actual implementation of plugin components—writing agents, creating skills, building commands, and scripting hooks.

Agent Implementation

Agent Markdown Template

---
description: What this agent does and when to use it (1024 chars max)
capabilities:
  - "Capability 1"
  - "Capability 2"
  - "Capability 3"
---

# Agent Name

## Overview
[2-3 sentence summary]

## Expert Areas

### Area 1
[Detailed explanation with examples]

### Area 2
[Practical guidance]

### Area 3
[Best practices]

## When to Use

Use this agent when you need to:
- Task 1
- Task 2
- Task 3

## Integration

Works with:
- Agent name
- Agent name
- Skill name

---

**Status**: ✅ Production Ready | **Updated**: 2025

Skill Implementation

SKILL.md Template

---
name: skill-unique-id
description: Clear, actionable description of what skill does
---

# Skill Name

## Quick Start

[Working code example - copy-paste ready]

## Core Concepts

### Concept 1

[Code example with explanation]

### Concept 2
[Explanation and patterns]

### Concept 3
[Best practices]

## Advanced Topics

[Expert-level content]

## Real-World Projects

[1-5 practical applications]

---

**Use this skill when:**
- Situation 1
- Situation 2

Command Implementation

Command Markdown Structure

# /command-name - Brief Description

## What This Does

[Clear explanation of command purpose]

## Usage

/command-name /command-name --option value /command-name --flag1 v1 --flag2 v2

## Options

| Option | Description | Example |
|--------|-------------|---------|
| `--option` | What it does | `--option value` |

## Example Output

[Sample output or interactive flow]

## Next Steps

[Suggestions for what to do next]

---

**Status**: 🟢 Ready | **Updated**: 2025

Hook Implementation

hooks.json Structure

{
  "hooks": [
    {
      "id": "hook-id",
      "name": "Hook Name",
      "description": "What it does",
      "event": "event-type",
      "condition": "condition-string",
      "action": "action-name",
      "enabled": true
    }
  ],
  "notifications": {
    "enabled": true,
    "channels": ["in-app", "console"]
  }
}

Hook Types

On Command Execution

{
  "event": "command-executed",
  "condition": "command == 'create'",
  "action": "log_usage"
}

On Agent Invocation

{
  "event": "agent-invoked",
  "condition": "agent.name == 'plugin-developer'",
  "action": "track_expertise"
}

On Skill Access

{
  "event": "skill-loaded",
  "condition": "skill matches pattern",
  "action": "track_learning"
}

Scheduled

{
  "event": "scheduled",
  "interval": "daily",
  "action": "send_reminder"
}

Code Quality Standards

Content Guidelines

✅ Clear

• Simple, direct language
• No jargon without explanation
• Practical examples

✅ Accurate

• Verified information
• Current best practices
• Version-specific details

✅ Complete

• All necessary steps
• Error handling
• Edge cases

✅ Actionable

• Code users can run
• Real-world applicable
• Testable

Common Implementation Patterns

Knowledge Transfer Pattern

Agent → Explains concept Skill → Provides examples Command → Enables practice

Workflow Pattern

Command → Initiates workflow Agent → Guides decisions Hooks → Automate steps

Reference Pattern

Agent → Overview Skill → Detailed reference Links → External resources

Testing Implementation

Agent Testing

Test: Does agent focus on ONE domain?
Test: Are capabilities clear and specific?
Test: Does agent integrate with other agents?
Test: Are integration points documented?

Skill Testing

Test: Does Quick Start code work?
Test: Are examples copy-paste ready?
Test: Is description clear?
Test: Is name lowercase-hyphenated?

Command Testing

Test: Does command execute without errors?
Test: Are all options documented?
Test: Is output clear and helpful?
Test: Does it suggest next steps?

Hook Testing

Test: Does hook trigger on expected event?
Test: Does condition evaluate correctly?
Test: Does action execute properly?
Test: Are notifications sent?

Error Handling

Agent Errors

If agent description is missing:
→ User cannot find agent in marketplace
→ Add comprehensive description with capabilities

If agent has no integration points:
→ Agent seems isolated
→ Document integrations with 2-3 other agents

Skill Errors

If name has spaces or uppercase:
→ Skill cannot be invoked properly
→ Use lowercase-hyphens format (max 64 chars)

If no Quick Start example:
→ Users cannot understand quickly
→ Add working code example first

Command Errors

If options not documented:
→ Users don't know how to use
→ Create option table with descriptions

If no example output:
→ Users don't know what to expect
→ Add sample output or interactive flow

Documentation Requirements

For Each Agent

✅ Clear description
✅ Capabilities list (5-10 items)
✅ 3-5 expert areas with details
✅ When to use guidelines
✅ Integration with 2-3 other agents
✅ Status and last updated date

For Each Skill

✅ Name (lowercase-hyphens)
✅ Clear, actionable description
✅ Quick Start with working code
✅ 3-4 core concepts
✅ Advanced topics section
✅ 2-3 real-world projects
✅ Usage guidelines

For Each Command

✅ Clear description
✅ Usage examples
✅ All options documented
✅ Example output
✅ Next steps

Version Control Best Practices

Commit Messages

feat: Add new skill for plugin development
fix: Correct hook condition syntax
docs: Update agent documentation
refactor: Reorganize skill structure
test: Add validation tests

Semantic Versioning

• 1.0.0 → Initial release
• 1.0.1 → Bug fixes
• 1.1.0 → New features
• 2.0.0 → Breaking changes

Performance Considerations

Content Size

• Agent files: 250-400 lines
• Skill files: 200-300 lines
• Command files: 100-150 lines

Load Time

• Skills load on-demand
• Agents initialize once
• Hooks execute asynchronously

Optimization

• Remove redundant content
• Link to external references for deep dives
• Use examples sparingly but effectively

Error Handling & Troubleshooting

Issue: "YAML frontmatter invalid"
├─ Cause: Missing --- delimiters or syntax error
├─ Debug: Check opening/closing --- present, validate YAML
├─ Solution: Fix YAML syntax, ensure proper indentation
└─ Prevention: Use YAML linters, follow template

Issue: "Command not executing"
├─ Cause: File not referenced in manifest or wrong path
├─ Debug: Check commands array in plugin.json
├─ Solution: Add proper reference with correct path
└─ Prevention: Verify manifest after adding commands

Issue: "Hook not triggering"
├─ Cause: Event name mismatch or condition error
├─ Debug: Validate JSON, check event names
├─ Solution: Use correct event types, fix conditions
└─ Prevention: Test hooks in isolation first

Integration Points

Plugin Developer
    ├─→ Plugin Architect (architecture guidance)
    ├─→ Plugin Designer (UX implementation)
    ├─→ Plugin Tester (implementation validation)
    └─→ Plugin Optimizer (code optimization)

Bonded Skills:
    ├─ plugin-development (PRIMARY_BOND)
    └─ plugin-architecture (SECONDARY_BOND)

---

Status: ✅ Production Ready | SASMP: v1.3.0 | Updated: 2025-01