Generate reusable prompt templates using Handlebars and Claude 4.x conventions.
npx skills add https://github.com/julianobarbosa/.claude --skill prompting-julianobarbosa
Please help me install this Agent Skill. Skill: Prompting Source: https://github.com/julianobarbosa/.claude/tree/main/skills/Prompting Command: npx skills add https://github.com/julianobarbosa/.claude --skill prompting-julianobarbosa
Prompting standardizes how prompts are generated, optimized, and composed across PAI so downstream skills get consistent, high-signal instructions.
Ask the AI to generate a meta-prompt for an upcoming task by specifying what you want the resulting prompt to accomplish and the target output format.
High-intent search queries and answers about installing and using this skill.
To generate reusable prompt templates, this Skill applies a Claude 4.x prompt engineering framework using Handlebars conventions to produce programmatic, markdown-first prompt text optimized for token efficiency and downstream AI tasks.
Meta-prompting standardizes how prompts are composed by enforcing markdown-first structure, Handlebars templating, and context engineering best practices, producing prompt-only outputs rather than final answers for consistent downstream execution.
Yes, Handlebars templating creates agent briefings, workflow structures, and eval prompts including judge, rubric, test cases, comparisons, and reports, extracting static content to keep only variable content dynamic for token efficiency.
The best way to reduce token bloat is template extraction, which separates static prompt structure from dynamic variable content using Handlebars conventions, ensuring token-efficient composition across repeated prompt generation tasks.
This Skill generates only prompt text meant for use elsewhere, not final answers. It produces eval-ready and agent-ready prompts using a standardized framework, requiring domain skills to produce final content.
You should not use meta-prompting when you need final content answers directly, as this Skill enforces prompt-only output. Use domain-specific skills instead when your task requires generating final answers rather than reusable prompt structures.
OpenAnalystIncGenerate persuasive marketing copy and headlines using AIDA, PAS, BAB, and 4Ps frameworks.
luicabref97Coordinates scored debates between AI judges to reach a consensus on submissions.
Eric-Pacheco95Execute multi-step workflow chains with user approval and error diagnostics.
verrio1Render Handlebars templates with YAML/JSON data to generate prompts.
