project-guide
Simulates plan mode with a teaching-first approach. Use when onboarding to a new codebase, exploring architecture, or seeking guided understanding of any project area. Produces high-level overviews with diagrams, drills into details on request, and offers follow-up exploration questions.
Install with Tessl CLI
npx tessl i github:0xrabbidfly/eric-cartman --skill project-guide85
Quality
83%
Does it follow best practices?
Impact
84%
1.33xAverage score across 3 eval scenarios
Evaluation results
94%
48%New Developer Onboarding: Project Orientation Guide
Phase 1 project overview format
Overview header
0%
50%
What it is field
0%
100%
Built with field
0%
100%
Runs on field
0%
100%
Folder structure diagram
100%
100%
Key entry points table
100%
100%
Follow-up questions section
40%
80%
At least 2 follow-up questions
0%
100%
Follow-up question explanations
0%
100%
No full file dumps
100%
100%
High-level first
100%
100%
At least one diagram
100%
100%
Without context: $0.2357 · 1m 12s · 9 turns · 14 in / 3,546 out tokens
With context: $0.6006 · 2m 21s · 22 turns · 59 in / 6,806 out tokens
90%
1%Architecture Documentation for Engineering Handbook
Architecture diagrams with Mermaid and ASCII
Mermaid diagram present
100%
100%
ASCII fallback present
100%
100%
System context level
100%
100%
Container/services level
100%
100%
Component level
100%
100%
Verb-labeled relationships
100%
100%
Directional arrows
100%
100%
Follow-up questions
0%
0%
No jargon without definition
70%
80%
High-level before detail
100%
100%
Without context: $0.2655 · 1m 42s · 10 turns · 15 in / 6,128 out tokens
With context: $0.5617 · 2m 7s · 17 turns · 2,760 in / 7,485 out tokens
69%
14%Codebase Deep Dive: Explaining a Complex Subsystem to a New Contributor
Deep dive format and complexity handling
Deep dive header
0%
37%
Where We Are section
100%
75%
Why It Exists section
62%
62%
How It Works section
75%
50%
Key Files table
100%
100%
Watch Out For section
50%
60%
Connections section
37%
100%
Complexity acknowledged
90%
40%
Simplified model first
30%
40%
Follow-up questions section
0%
87%
No raw file dumps
100%
100%
Follow-up italic rationale
0%
100%
Without context: $0.2874 · 1m 46s · 13 turns · 20 in / 5,201 out tokens
With context: $0.5566 · 2m 15s · 19 turns · 23 in / 6,734 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.