Key Findings

Discovered Topics

๐Ÿ”จ
Project Structure
Build commands, package managers, and dependency setup
task
0.4688
git
0.3965
commit
0.3554
behavior
0.3259
repository
0.3251
check
0.3198
work
0.3158
doc
0.2874
project
0.2769
workflow
0.2761

This topic represents a cluster of related terms discovered through topic modeling analysis. The word weights indicate how strongly each term is associated with this theme across the analyzed repositories.

๐Ÿงช
Frontend Development
Test runners, linting, and validation workflows
npm
1.7814
typescript
0.3029
frontend
0.2997
react
0.2771
lint
0.2241
api
0.212
pnpm
0.2086
component
0.1896
typecheck
0.1895
dev
0.1691

Frontend-focused CLAUDE.md files are among the most detailed. The prevalence of "npm," "typescript," and "react" reflects the dominance of the React/TypeScript ecosystem in projects using Claude Code. These files typically specify component patterns, state management, and build tooling.

โœ๏ธ
Nextjs & Training
Code style, formatting conventions, and naming rules
nextjs
0.6259
training
0.6196
breaking
0.5497
writing
0.5458
know
0.525
apis
0.4721
guide
0.4337
relevant
0.4273
version
0.3779
structure
0.3776

This topic represents a cluster of related terms discovered through topic modeling analysis. The word weights indicate how strongly each term is associated with this theme across the analyzed repositories.

๐Ÿ›๏ธ
Tool & Import
Project architecture, modules, and directory structure
tool
0.6218
import
0.3567
model
0.2904
return
0.284
type
0.2263
error
0.2165
mcp
0.2068
function
0.203
const
0.202
string
0.2012

This topic represents a cluster of related terms discovered through topic modeling analysis. The word weights indicate how strongly each term is associated with this theme across the analyzed repositories.

โ›”
Skill & Para
Constraints, prohibitions, and agent guardrails
skill
1.3295
para
0.2126
codex
0.205
del
0.1811
issue
0.1785
markdown
0.1645
claude
0.1491
frontmatter
0.142
domain
0.1355
instruction
0.1305

This topic represents a cluster of related terms discovered through topic modeling analysis. The word weights indicate how strongly each term is associated with this theme across the analyzed repositories.