skills.rest
robinlllll
robinlllllCOMMUNITY
Explore all skills by this author →
GitHub

prompt-optimizer

Develop and refine prompts using a PromptSpec-driven workflow with tests and versioning.

Updated Aug 23, 2026
Search Tags:#workflow#versioning#prompt-engineering#ai-review#prompt-optimization
One-click install
npx skills add https://github.com/robinlllll/claude-code-skills --skill prompt-optimizer-robinlllll
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-optimizer
Source: https://github.com/robinlllll/claude-code-skills/tree/main/prompt-optimizer
Command: npx skills add https://github.com/robinlllll/claude-code-skills --skill prompt-optimizer-robinlllll

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai>=1.0.0, openai>=1.0.0, python-dotenv>=1.0.0, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables teams to iteratively craft, test, and version prompts by enforcing a formal PromptSpec, test suites, and a transparent review process.

Core Features & Use Cases

  • Define and enforce a PromptSpec to guide prompt behavior
  • Run deterministic tests via TestSuite and track changes across versions
  • Use multi-AI review (Gemini + GPT) to audit prompts in parallel
  • Mandatory user testing before iterations to ensure real-world viability
  • Manage resources (scripts, references, assets) for on-demand operations as described in SKILL.md

Quick Start

Create a new session with a clear goal, populate spec.md, add an initial version, and run a dual AI review to iterate.

Frequently Asked Questions about prompt-optimizer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I iteratively optimize prompts with AI review?▼

Iterative prompt optimization applies a formal PromptSpec workflow with tests and versioning, running parallel Gemini and GPT reviews to audit and track prompt improvements.

What is a PromptSpec and how does it guide prompt behavior?▼

A PromptSpec is a formal specification used to define and enforce prompt behavior, ensuring that prompt engineering follows a structured, testable workflow rather than ad-hoc changes.

How do I track prompt versions and run tests during prompt engineering?▼

You track prompt versions by running deterministic tests via the TestSuite, which logs changes across versions to ensure updates maintain real-world viability before iterations.

Does prompt optimization work with both OpenAI and Gemini for parallel AI review?▼

Yes, prompt optimization supports parallel multi-AI review using both OpenAI and Gemini dependencies to audit prompts simultaneously and identify improvements from multiple models.

What is the best way to start a prompt optimization session?▼

The best way to start is to create a new session with a clear goal, populate spec.md, add an initial prompt version, and run a dual AI review to begin iterating.

When do I need formal versioning for prompt engineering?▼

You need formal versioning for prompt engineering when managing complex changes requires deterministic tests and transparent tracking to ensure real-world viability across iterations.

Related Skills

Explore More Product & Management Skills→
trungdo9trungdo9
3

ai-multimodal

Process and generate multimedia content using the Google Gemini API.

Community
Advanced
andregusman-raizandregusman-raiz
19

ag-publicar-deploy

Deploy code to preview, production, or rollback environments with automated smoke tests.

Community
Advanced
truchottruchot
2

testing-process

Define testing strategies and orchestrate test routing across CI/CD workflows.

Community
Advanced
SheikhMuhammadHamza-HSSheikhMuhammadHamza-HS
1

mcp

Integrate AI applications with external systems via the Model Context Protocol.

Community
Advanced