oc-prompt-ops

Manage prompt versioning, evaluation, and drift detection for LLMs.

Updated May 15, 2026
One-click install
npx skills add https://github.com/asfbay-bit/opchain-skills --skill oc-prompt-ops
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oc-prompt-ops
Source: https://github.com/asfbay-bit/opchain-skills/tree/main/skills/oc-prompt-ops
Command: npx skills add https://github.com/asfbay-bit/opchain-skills --skill oc-prompt-ops

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires oc-claude-api, oc-agent-forge, oc-rag-forge, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill manages the versioning, evaluation, and drift detection of prompts for Large Language Models (LLMs), ensuring consistent quality and reducing manual effort.

Core Features & Use Cases

  • Prompt Versioning: Treat prompts as source-controlled code with versioning and diffing capabilities.
  • Evaluation: Run prompts against predefined datasets to evaluate their performance.
  • Drift Detection: Monitor and detect changes in prompt performance due to model updates or external factors.
  • Use Case: When developing an LLM application, use this Skill to manage the lifecycle of your prompts, ensuring they remain effective over time.

Quick Start

Run the command '/oc-prompt eval [email protected]' to evaluate the current version of the 'model-routing' prompt against the 'v1.2.0' dataset.

Frequently Asked Questions about oc-prompt-ops

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

FAQPage Schema
How do I manage prompt versioning for LLM applications?

Prompt versioning for LLM applications treats prompts as source-controlled code with versioning and diffing capabilities. This ensures consistent quality and reduces manual effort during prompt lifecycle management.

What is LLM prompt drift detection and when do I need it?

LLM prompt drift detection monitors and identifies changes in prompt performance caused by model updates or external factors. You need it to ensure your prompts remain effective and consistently produce expected outputs over time.

How do I evaluate LLM prompts against predefined datasets?

Evaluate LLM prompts by running specific prompt versions against predefined datasets to measure their performance. This process identifies quality changes and ensures prompts meet expected output standards.

Do I need dependencies to run prompt evaluation and drift detection?

Yes, full functionality for prompt evaluation and drift detection requires dependencies including the Claude API, agent forge, and RAG forge components. These provide the necessary environment for complete prompt lifecycle management.

What's the best way to track prompt changes during LLM model updates?

The best way to track prompt changes during LLM model updates is using prompt versioning combined with drift detection. This monitors performance shifts caused by model updates and maintains prompt effectiveness through diffing capabilities.