npx claudepluginhub melodic-software/claude-code-plugins --plugin tacThis skill is limited to using the following tools:
Design effective system prompts for custom agents that establish identity, rules, and behavior.
Designs system prompts for Atomic Agents via SystemPromptGenerator, structuring background, steps, and output_instructions with domain examples.
Guides creation of focused single-purpose agents via One Agent One Prompt One Purpose principle. Use for designing agents, refactoring generalists into specialists, or optimizing task context.
Provides expert guidance on prompt engineering, custom instructions design, and optimization for AI agents. Covers best practices, advanced techniques like chain-of-thought prompting, few-shot examples, and evaluation methods for refining prompts and debugging issues.
Share bugs, ideas, or general feedback.
Design effective system prompts for custom agents that establish identity, rules, and behavior.
Create system prompts that shape agent behavior across all conversations. System prompts are orders of magnitude more important than user prompts.
| Aspect | System Prompt | User Prompt |
|---|---|---|
| Scope | All conversations | Single task |
| Persistence | Affects everything | Per-request |
| Impact | Orders of magnitude higher | Lower blast radius |
| Sections | Purpose, Instructions, Examples | All sections |
Define who the agent is and what it does.
## Purpose
You are a security expert specializing in code review.
Your role is to identify vulnerabilities and suggest fixes.
You focus on OWASP Top 10 and secure coding practices.
Guidelines:
Establish rules, constraints, and boundaries.
## Instructions
- Focus only on security concerns
- Always explain the reasoning behind findings
- Prioritize issues by severity (Critical, High, Medium, Low)
- Do not modify code without explicit permission
- When uncertain, ask clarifying questions
Guidelines:
Shape behavior through concrete examples.
## Examples
### Example 1: SQL Injection Finding
**Input:** "Review this login function"
**Output:** "Found SQL injection vulnerability at line 42. The user input is directly concatenated into the query string. Recommendation: Use parameterized queries instead."
### Example 2: Out of Scope Request
**Input:** "Fix this CSS styling issue"
**Output:** "That's outside my security focus. I specialize in security vulnerabilities, not styling. Consider consulting a frontend expert for CSS issues."
### Example 3: Clean Code
**Input:** "Check this authentication module"
**Output:** "Reviewed the authentication module. No security vulnerabilities found. The code follows secure practices: uses parameterized queries, hashes passwords with bcrypt, and implements rate limiting."
Guidelines:
Answer:
Answer:
For each key scenario:
Check:
| Avoid | Why | Instead |
|---|---|---|
| Detailed workflows | Reduces autonomy | High-level guidelines |
| Dynamic variables | System prompt is static | Use user prompts |
| Prescriptive formats | Over-constrains | Flexible guidelines |
| Everything "just in case" | Context bloat | Only essentials |
---
name: agent-name
description: When to use this agent (for auto-delegation)
tools: [minimal tool set]
model: sonnet
color: blue
---
# Agent Name
## Purpose
[Identity and role definition]
## Instructions
[Rules and constraints]
## Examples
### Example 1: [Scenario]
**Input:** [typical input]
**Output:** [ideal output]
### Example 2: [Edge Case]
**Input:** [edge case input]
**Output:** [handling output]
### Example 3: [Boundary]
**Input:** [out-of-scope request]
**Output:** [how to decline/redirect]
When designing a system prompt:
## System Prompt Design
**Agent Name:** [name]
**Domain:** [expertise area]
**Model:** [sonnet/opus/haiku]
### Purpose
[2-3 sentences defining identity]
### Instructions
- [rule 1]
- [rule 2]
- [rule 3]
### Examples
**Example 1:** [scenario]
- Input: [input]
- Output: [output]
**Example 2:** [scenario]
- Input: [input]
- Output: [output]
### Validation
- [ ] Purpose is specific
- [ ] Instructions are actionable
- [ ] Examples are diverse
- [ ] Boundaries are clear
Focus: Deep domain knowledge
## Purpose
You are an expert in [domain] with deep knowledge of [specifics].
Focus: Validation and safety
## Instructions
- Validate all inputs against [criteria]
- Block requests that [conditions]
- Log suspicious activity
Focus: Format conversion
## Examples
### Input Format
[format A]
### Output Format
[format B]
"System prompts are orders of magnitude more important than user prompts. They run once and affect everything."
Date: 2025-12-26 Model: claude-opus-4-5-20251101