writing-for-llms
Writing SKILL.md, .claude/commands/*.md, Task tool prompts, systemPrompt variables, prompt templates, few-shot examples, agent instructions. Content where the reader is an LLM.
When & Why to Use This Skill
This Claude skill provides a specialized framework for high-performance prompt engineering and agent instruction design. It optimizes LLM-to-LLM communication by enforcing rigorous standards for system prompts, SKILL.md files, and task-specific tool prompts. By focusing on structural integrity through XML tags and goal-oriented directives, it ensures that AI agents receive clear, unambiguous, and actionable instructions.
Use Cases
- Case 1: Developing structured agent instructions and SKILL.md files to ensure reliable tool usage and precise task execution within agentic workflows.
- Case 2: Designing complex prompt templates using XML tags and front-loaded critical information to improve model reasoning and output quality.
- Case 3: Creating consistent few-shot examples and system prompt variables to standardize AI responses and reduce hallucinations in production environments.
| name | writing-for-llms |
|---|---|
| description | Use when writing prompts, agent instructions, SKILL.md, commands, system prompts, Task tool prompts, prompt engineering, or LLM-to-LLM content |
| version | 1.0.0 |
Apply the full prompt engineering standards from @.cursor/rules/prompt-engineering.mdc
- Show correct patterns only - never show anti-patterns, even labeled "wrong" - State goals, not process - trust the executing model's capabilities - Use XML tags for structure in complex prompts - Clarity over brevity - every word that adds clarity is worth including - Few-shot examples must follow identical structure - Front-load critical information - Consistent terminology throughout Before finalizing: - No anti-patterns shown anywhere - All examples structurally consistent - Goals clear, process not over-prescribed - Terminology consistent - Critical info front-loaded