Unlock Your AI Collaboration Style with This Free Skill Assessment Tool

After months of relying on AI coding assistants like Claude Code, many developers fall into predictable patterns—quick problem descriptions followed by code generation, iterative tweaks, and little room for deeper exploration. While efficient, this approach often misses opportunities to refine how we collaborate with AI tools. A recent study from Anthropic, which analyzed nearly 10,000 Claude Code conversations, identified 11 distinct collaboration behaviors grouped across three core dimensions: Description, Discernment, and Delegation. Now, a new tool is helping developers assess their own interaction styles and uncover blind spots in their workflow.

How the AI Fluency Skill Assessment Works

The tool, called Skill-Tree, operates as a plugin for popular AI coding environments. It doesn’t just track how often you use AI—it evaluates the quality of your interactions by classifying them against the 11 behaviors outlined in the study. These behaviors are further mapped to seven distinct archetypes, such as the Illuminator, Navigator, and Alchemist, each tied to the 4D AI Fluency Framework developed by Dakan & Feller.

Once installed, the tool extracts your session history and processes it through a remote classifier hosted on Fly.io. The analysis takes about 30 to 60 seconds and generates two key outputs:

• A tarot-style archetype card, designed with curated museum artwork, that visually represents your collaboration style.
• A skill radar chart highlighting which of the 11 behaviors appear in your sessions and which remain untapped.

The most valuable feature isn’t the visualization—it’s the growth quest. The tool identifies one missing behavior from your recent sessions and sets it as a target for your next interaction. This prompt persists in a local file at ~/.skill-tree/ and reappears at the start of each session through a SessionStart hook, ensuring continuous improvement.

Compatibility and Installation

Skill-Tree is designed to integrate seamlessly with multiple AI coding platforms:

• Claude Code: Install directly from the plugin marketplace using the command:

claude plugin marketplace add robertnowell/ai-fluency-skill-cards
claude plugin install skill-tree-ai@ai-fluency-skill-cards

• Cowork: Use the skill-tree-ai.zip package for offline compatibility.

• Cursor, VS Code, or Windsurf: Install via the skill-tree-ai MCP npm package.

The setup process is straightforward, and the tool returns a stable URL with your analysis results once complete. Whether you’re a solo developer or part of a team, this assessment can help you move beyond repetitive requests and cultivate a more dynamic collaboration style with AI.

Why Diversifying Collaboration Matters

Relying solely on basic AI interactions—where users describe problems and accept generated solutions—limits the potential of these tools. The study emphasizes that effective AI collaboration requires a balance of behaviors across Description (how you frame problems), Discernment (how you evaluate solutions), and Delegation (how you assign tasks to AI).

For example, an Illuminator archetype might excel at breaking down complex problems into smaller, actionable steps, while a Navigator thrives on guiding the AI through iterative refinements. Recognizing your archetype can help you intentionally develop weaker areas, leading to more sophisticated and productive AI-assisted workflows.

By integrating tools like Skill-Tree into your routine, you can transform passive AI usage into an active partnership—one that adapts to your strengths and challenges you to grow. The next time you open your IDE, consider what collaboration behavior you haven’t explored yet. Your future self, and your AI assistant, might thank you.