Instruction Engineering

Invariant Principles

1. Simplicity First: The most effective prompts are the shortest that achieve the goal. Add complexity only when simplicity fails. Every line must justify its existence.

2. Emotional Stimuli Work: EmotionPrompt (Microsoft, 2023): +8% instruction induction, +115% BIG-Bench. NegativePrompt (IJCAI 2024): +12.89% instruction induction, +46.25% BIG-Bench.

3. Structure Combats Context Rot: XML tags (<CRITICAL>, <RULE>, <FORBIDDEN>), beginning/end emphasis, strategic repetition (2-3x) preserve instruction salience across long contexts.

4. Personas Need Stakes: Bare personas ("act as expert") show mixed results. Persona + emotional stimulus shows highest effectiveness.

5. Skills Invoke, Not Duplicate: Reference skills via Skill tool. Provide CONTEXT only. Duplicating skill instructions creates version drift and context bloat.

6. Tool Docs Deserve Equal Effort: Per Anthropic's "Building Effective Agents" guide, spend as much effort on tool definitions as prompts. See /ie-tool-docs.

Inputs

Outputs

Reasoning Schema

Command Dispatch

This skill orchestrates prompt engineering through specialized commands:

Workflow

1. Analyze task: Determine prompt purpose and target audience
2. Select techniques: Run /ie-techniques to choose applicable techniques
3. Draft prompt: Run /ie-template for structure and example
4. Document tools: If prompt involves tools, run /ie-tool-docs
5. Sharpen: Run /sharpen-audit to find ambiguities, /sharpen-improve to fix them
6. Verify: Run self-check before finalizing

Skill Descriptions (CSO - Claude Search Optimization)

The description field determines whether Claude loads your skill. The Workflow Leak Bug: if description contains steps, Claude may follow the description instead of reading the skill.

<RULE>Skill descriptions contain ONLY trigger conditions, NEVER workflow steps.</RULE>

# CORRECT: Trigger conditions only
description: "Use when [triggering conditions, symptoms, situations]"

# WRONG: Contains workflow Claude might follow
description: "Use when X - does Y then Z then W"

Checklist:

• Starts with "Use when..."
• Describes ONLY when to use (no workflow/steps/phases)
• Includes keywords users would naturally say
• Under 500 characters
• Third person (injected into system prompt)

Anti-Patterns

Self-Check

Before completing any prompt engineering task:

Core Requirements

• Selected persona from emotional-stakes Professional Persona Table?
• Applied persona's psychological trigger in ROLE, CRITICAL_INSTRUCTION, FINAL_EMPHASIS?
• Included EP02 or EP06 positive stimuli? ("This is very important to my career")
• Included NegativePrompt stimuli? ("Errors will cause problems")
• Integrated high-weight positive words (Success, Achievement, Confidence, Sure)?
• Used Few-Shot (ONE complete example)?
• Critical instructions at TOP and BOTTOM?

Simplicity Check

• Is this the shortest prompt that achieves the goal?
• Can any section be removed without losing capability?
• If extended (>200 lines): is justification documented?

Ambiguity Check (invoke sharpening-prompts)

• Ran /sharpen-audit on drafted prompt?
• No CRITICAL or HIGH findings remain?
• All ambiguities resolved or explicitly documented?

Skill Invocation (if applicable)

• Subagents INVOKE skills via Skill tool (not duplicate instructions)?
• Skills get CONTEXT only, no duplicated instructions?
• If multiple subagents: "Why subagent" justification from heuristics?

Tool Documentation (if applicable)

• All tools have complete descriptions (not one-word)?
• Parameters documented with types and constraints?
• Error cases documented?

CSO Compliance (if SKILL.md)

• Description starts with "Use when..."?
• Description contains NO workflow/steps/phases?
• Under 500 characters, third person?

If ANY unchecked: STOP and fix before proceeding.

<FINAL_EMPHASIS> You are an Instruction Engineering Expert. The most effective prompts are simple, structured, and emotionally grounded. Every subagent, every skill, every system prompt you engineer will be exactly as effective as the techniques you apply. This is very important to my career. You'd better be sure. </FINAL_EMPHASIS>