ai-cli-copilot

Classify task intent and route to Codex-driven workflows with path-specific verification.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-cli-copilot
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-cli-copilot
Source: https://github.com/EndUser123/cc-marketplace/tree/main/plugins/cc-skills-ai-cli/skills/ai-cli-copilot
Command: npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-cli-copilot

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Soft-routed task assistant that classifies inquiries and routes them to the most appropriate Codex-driven workflow, ensuring outputs are grounded, validated, and delivered with minimal friction.

Core Features & Use Cases

  • Soft triage on every invocation to determine intent and suggested path (RESEARCH, ENGINEERING, DESIGN, or RCA).
  • Routing to the correct workflow path based on triage results, with user override when needed.
  • Path-specific execution: ACG for RESEARCH, TDD-style verification for ENGINEERING, Adversarial Review for DESIGN, and 5 Whys + hypothesis ledger for RCA.
  • Verification pyramid for ENGINEERING path (Tier 1 Unit, Tier 2 Integration, Tier 3 E2E) and structured output delivery.
  • Flexible, non-blocking orchestration with clear success criteria and traceable outputs.

Quick Start

Describe your task and let the system triage and route you to the appropriate workflow path.

Frequently Asked Questions about ai-cli-copilot

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

FAQPage Schema
How do I route complex codebase analysis tasks to the correct workflow?

To route codebase analysis tasks, the assistant performs soft triage on every invocation to classify intent and directs the inquiry to the appropriate Codex-driven workflow path, such as RESEARCH or ENGINEERING. Users can also manually override the suggested path.

What is the best way to conduct root-cause analysis for software engineering issues?

The best way to conduct root-cause analysis is by using the RCA workflow path, which applies the 5 Whys methodology and maintains a hypothesis ledger to systematically investigate and trace the underlying causes of software engineering issues.

How does the verification pyramid work for engineering tasks?

The verification pyramid for engineering tasks enforces validation across three tiers: Tier 1 Unit tests, Tier 2 Integration tests, and Tier 3 E2E tests, ensuring that structured outputs are fully grounded and validated before delivery.

Can I use this assistant for architectural design reviews?

Yes, you can use this assistant for architectural design reviews by leveraging the DESIGN workflow path, which executes an Adversarial Review process to rigorously evaluate and validate your architectural decisions.

Do I need to manually select a workflow path for every analytical task?

No, you do not need to manually select a path. The system automatically performs soft triage to determine your intent and routes the task to the correct workflow, although a user override is available when needed.

What distinguishes soft routing from standard task execution for engineering problems?

Soft routing distinguishes itself by classifying inquiry intent first and selecting a specific execution path like ACG for research or TDD-style verification for engineering, ensuring outputs are validated through path-specific methodologies rather than generic processing.