AgentSkill: Semantic Anchor Translator

The Semantic Anchors project includes an AgentSkill that teaches AI coding assistants to recognize and suggest semantic anchors during your work.

What does it do?

The skill operates in two modes:

Recognition Mode (verbose → anchor)

When you describe a concept in your own words, the AI identifies the matching semantic anchor:

You: "I want to categorize things so they don’t overlap and cover everything"
AI: You’re describing MECE Principle (Mutually Exclusive, Collectively Exhaustive).

Guidance Mode (question → anchors)

When you ask for methodology advice, the AI suggests relevant anchors:

You: "How should I document architecture decisions?"
AI: For architecture decisions: ADR according to Nygard or MADR. Both provide lightweight templates for capturing context, decision, and consequences.

Installation

The skill follows the agentskills.io specification and can be used with any compatible AI coding assistant.

Claude Code

Add to your project’s CLAUDE.md:

## Skills

Use the semantic-anchor-translator skill from:
https://github.com/LLM-Coding/Semantic-Anchors/tree/main/skill/semantic-anchor-translator

Or reference the skill’s SKILL.md directly in your project configuration.

GitHub Copilot

Add to .github/copilot-instructions.md in your repository:

## Semantic Anchors

When the user describes a concept verbosely, identify the precise
semantic anchor term. When asked for methodology advice, suggest
relevant anchors from the catalog.

Reference: https://github.com/LLM-Coding/Semantic-Anchors/blob/main/skill/semantic-anchor-translator/references/catalog.md

Cursor

Add to .cursor/rules or .cursorrules in your project:

## Semantic Anchors

Use the semantic-anchor-translator skill to recognize and suggest
established terminology. Catalog:
https://github.com/LLM-Coding/Semantic-Anchors/blob/main/skill/semantic-anchor-translator/references/catalog.md

Amazon Kiro

Add to your project’s specs/ directory or include in a spec file:

## Semantic Anchors

When the user describes a concept verbosely, identify the precise
semantic anchor term. When asked for methodology advice, suggest
relevant anchors from the catalog.

Reference: https://github.com/LLM-Coding/Semantic-Anchors/blob/main/skill/semantic-anchor-translator/references/catalog.md

Other AI Tools

Any AI coding assistant that supports custom instructions or system prompts can use the skill. Point it to the catalog at:

https://github.com/LLM-Coding/Semantic-Anchors/blob/main/skill/semantic-anchor-translator/SKILL.md

Catalog Overview

The skill covers 90+ semantic anchors across these categories:

Category Key Anchors

Testing & Quality

TDD Chicago/London, BDD, Testing Pyramid, Mutation Testing, Property-Based Testing, Test Double

Software Architecture

Clean Architecture, Hexagonal, DDD, arc42, C4, ADR, MADR, CQRS, EDA

Design Principles

SOLID (5 sub-principles), GoF Design Patterns (23 sub-patterns), DRY, SPOT, YAGNI

Problem-Solving

Five Whys, Feynman Technique, Rubber Duck Debugging, Devil’s Advocate, Cynefin

Requirements Engineering

MoSCoW, EARS, User Story Mapping, JTBD, Impact Mapping

Communication

BLUF, Pyramid Principle, MECE, Chatham House Rule, Socratic Method

Documentation

Diataxis Framework, Docs-as-Code

Strategic Planning

Wardley Mapping, Pugh Matrix

The full catalog with descriptions, proponents, and core concepts is available at catalog.md on GitHub.

Contributing

When you add a new anchor to the catalog, please also update the AgentSkill catalog at skill/semantic-anchor-translator/references/catalog.md so AI agents can discover it.

See Contributing for the full contribution workflow.

See also

  • Socratic Code-Theory Recovery Skill — packages the brownfield documentation-recovery workflow as an installable skill that classifies every claim as code-derivable or open-to-the-team