Render Handlebars templates into prompts from YAML data.
npx skills add https://github.com/Steffen025/pai-opencode --skill prompting-steffen025
Please help me install this Agent Skill. Skill: Prompting Source: https://github.com/Steffen025/pai-opencode/tree/main/.opencode/skills/Prompting Command: npx skills add https://github.com/Steffen025/pai-opencode --skill prompting-steffen025
This skill enables systematic meta-prompting and template generation to standardize and accelerate prompt engineering.
To generate a meta-prompt, provide a task description and let the system assemble a target prompt using the Ultimate Prompt Template. For example, ask: "Create a structured research briefing for evaluating multiple sources on climate policy." Then render with the template engine: bun run RenderTemplate.ts --template Primitives/Briefing.hbs --data Data/Agents.yaml --output Compiled/MetaPrompt.md
High-intent search queries and answers about installing and using this skill.
Meta-prompting is a software-driven approach that generates prompts capable of writing other prompts. It standardizes prompt engineering by using a Handlebars-based templating system and a canonical frontmatter schema to assemble dynamic prompts for multi-context AI agent workflows.
To generate and render dynamic prompts from YAML data, use a rendering tool to assemble templates with structured data. Execute the rendering script by specifying the Handlebars template file, the YAML data source, and the output markdown file path to compile the final prompt.
Yes, you can use Handlebars templates to build multi-context prompts across various AI providers. The system applies a standardized, software-driven approach that relies on reusable prompt primitives to compose robust, provider-agnostic prompts for dynamic agent workflows.
The best way to structure reusable prompt primitives is to utilize a templates library containing components like Roster, Voice, Structure, Briefing, and Gate. These primitives allow you to systematically compose robust prompts for tasks such as analyzing sources and creating structured research briefs.
Yes, template-based prompt generation validates the structure of assembled prompts. The rendering and validation process includes built-in checks that verify the compiled prompt against the canonical frontmatter schema, significantly reducing structural errors before deployment.
windmill-labsCreate Bun/TypeScript scripts with an async main function template.
mathiasbourgoinDecompose validated research briefs into role-specific sub-briefs for execution agents.
githubnextGenerate structured task descriptions for GitHub Copilot agents using gh-aw frontmatter conventions.
starlink-awakenGenerate and compose prompts using Handlebars templates with YAML or JSON data.
