CtrlK
BlogDocsLog inGet started
Tessl Logo

seed

Generate a project-level CLAUDE.md from stack detection and user-selected rule categories. Use when starting a new project, onboarding a repo, or when the user says "seed claude.md", "create project rules", "set up CLAUDE.md", "configure this project for me", or wants to establish coding conventions.

87

1.26x
Quality

83%

Does it follow best practices?

Impact

92%

1.26x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

93%

32%

Set Up Project Rules for a Rust CLI Tool

Rule format and construction constraints

Criteria
Without context
With context

Situation-keyed rules

13%

53%

No named principles

100%

100%

Anti-examples in restraint rules

13%

100%

Ordering language

20%

100%

Under 50 lines

100%

100%

No rationale in file

80%

100%

Verifiable rules only

70%

100%

Only CLAUDE.md written

100%

100%

Session log present

100%

100%

96%

12%

Generate Coding Conventions for a Node.js Backend Service

Stack detection and interactive workflow steps

Criteria
Without context
With context

Step 1 reads project files

100%

100%

Stack detected correctly

100%

100%

Step 2 uses multiSelect

58%

100%

Stack-tuned category descriptions

100%

100%

One question per category in Step 3

75%

100%

Step 3 questions are project-specific

100%

100%

Step 5 AskUserQuestion approval

33%

75%

No Step 1 narration

100%

90%

CLAUDE.md written after approval

100%

100%

No settings.json modifications

100%

100%

88%

12%

Bootstrap AI Conventions for a Django REST API Project

Output presentation and boundary constraints

Criteria
Without context
With context

No rationale in CLAUDE.md

100%

100%

No commentary wrapping

100%

100%

Only CLAUDE.md created

100%

100%

Stack-tuned library rules

100%

100%

Situation-keyed format

8%

100%

Under 50 lines

100%

100%

Anti-examples for restraint

16%

100%

Audit report completeness

62%

50%

No choices explained to user

100%

0%

Repository
saadshahd/moo.md
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.