Classify user request intent and generate Prometheus directives with acceptance criteria.
npx skills add https://github.com/Lee-SiHyeon/oh-my-copilot --skill metis-lee-sihyeon
Please help me install this Agent Skill. Skill: metis Source: https://github.com/Lee-SiHyeon/oh-my-copilot/tree/main/.github/skills/metis Command: npx skills add https://github.com/Lee-SiHyeon/oh-my-copilot --skill metis-lee-sihyeon
Metis reduces costly AI failures by classifying the user’s intent up front and surfacing hidden ambiguities, mismatches, and likely failure points before any planning or execution begins.
Ask Metis to classify your request intent and produce Prometheus directives and executable acceptance criteria for the planned work.
High-intent search queries and answers about installing and using this skill.
Prevent AI planning failures by using intent-aware pre-planning to classify your refactoring request and surface hidden ambiguities. This upfront analysis generates targeted probing questions and executable acceptance criteria, ensuring boundaries are clarified before execution begins.
Intent classification for AI quality guardrails determines whether a request is Refactoring, Build from Scratch, Mid-sized Task, Collaborative, Architecture, or Research. It applies intent-specific safety rules to enforce boundaries and prevent misfires during software architecture planning.
Generate executable acceptance criteria for a greenfield build by classifying the intent and producing structured Prometheus directives. These directives define explicit MUST and MUST NOT constraints, prioritizing executable QA and anti-slop measures for the planned build.
Yes, you can use pre-planning to clarify scope for mid-sized delivery tasks. It applies intent-specific analysis to ask precise questions tailored to your task type, clarifying boundaries and acceptance criteria without relying on vague scope prompts.
No, this pre-planning approach does not modify code directly during architecture decisions. It enforces a read-only advisory workflow that analyzes user requests, classifies intent, and outputs structured directives and probing questions only.
The best way to define must-have and exclusion constraints is by generating Prometheus directives. These structured outputs provide explicit MUST and MUST NOT rules that prioritize executable QA, enforce explicit exclusions, and apply anti-slop measures for AI development.
flosrnRoute marketing tasks to optimal specialized skills based on user intent.
jaisonerickImplement story files into tested code and update execution logs.
kunalrawat425Evaluate architecture and plan quality for engineering initiatives.
amadeus-dlcExecute an isolated AI-DLC code-generation stage without advancing the main workflow.
