What files does it check for project context?

It specifically looks for .claude/workspace-info.toon, CLAUDE.md, README.md, and relevant configuration files like package.json, tsconfig.json, or pyproject.toml.

Does it support external documentation?

Yes, it can utilize available MCP documentation tools and perform targeted web searches for official API references and library documentation when local information is insufficient.

How does this skill decide whether to search the web or ask me?

It classifies the ambiguity first: technical facts trigger a tiered search starting with local files and moving to the web, while questions about your specific intent or preference go directly to you.

Why use this instead of just letting Claude guess?

Guessing often leads to subtle bugs and rework. This skill enforces a 'rather ask than guess' principle, ensuring that the implementation matches your project's actual requirements and standards.

Smart Ambiguity Resolver

Name: Smart Ambiguity Resolver
Author: rayk

byrayk

•

Productivity & Workflow

Eliminates guesswork in development by implementing a systematic, tiered approach to gathering missing information and requirements.

The Smart Ambiguity Resolver is a specialized Claude Code skill designed to minimize wasted development cycles caused by incorrect assumptions. By following a rigorous decision tree, it classifies ambiguities as either technical facts or user intent, then systematically scans project context, architectural documentation, and web sources before finally prompting the user with structured choices. It ensures that every action Claude takes is grounded in authoritative data, significantly reducing bugs and misaligned implementations in complex codebases.

Key Features

01Tiered lookup process prioritizing local project context over web searches

02Automated scanning of .toon context files and architectural documentation

03Structured user clarification with prioritized options and trade-off explanations

041 GitHub stars

05Intelligent classification between technical questions and intent-based choices

06Integration with MCP documentation tools for high-fidelity technical lookups

Use Cases

01Determining project-specific naming conventions and architectural patterns

02Clarifying specific API syntax or library versions during implementation

03Choosing between multiple valid implementation strategies based on user preference

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rayk/lucid-toolkit resolve-ambiguity

For use in Claude.ai and ChatGPT

Download Skill