ebroky/nsfw

View on GitHub by @ebroky

🔍 Detect NSFW content with this lightweight HTTP API. Ideal for CPU machines, it identifies sensitive images quickly and efficiently.

What's novel

🔍 Detect NSFW content with this lightweight HTTP API. Ideal for CPU machines, it identifies sensitive images quickly and efficiently.

Code Analysis

13 files read · 4 rounds

A FastAPI HTTP service that wraps a pre-trained ONNX NSFW detection model, accepting image URLs or file uploads, running inference, and returning SFW/NSFW probability scores.

Strengths

The ONNX model is loaded once at startup for performance, and the image preprocessing pipeline (resize, normalize, GIF frame extraction) is reasonably well-implemented with proper edge case handling for different image formats.

Weaknesses

Massive code duplication between check() and post_check(), significant dead code (rate limiting, DB models, email validation), no tests, and the README describes a completely different product (desktop app with .exe/.dmg) than what the code actually is (Docker-based Python API service).

Score Breakdown

Innovation

2 (25%)

Craft

34 (35%)

Traction

14 (15%)

Scope

38 (25%)

Signal breakdown

Innovation

Not Fork+1

Code Novelty+0

Concept Novelty+0

Craft

Ci-3

Tests-5

Polish+0

Releases-2

Has License+5

Code Quality+13

Readme Quality+15

Recent Activity+7

Structure Quality+5

Commit Consistency+2

Has Dependency Mgmt+5

Traction

Forks+6

Stars+6

Hn Points+0

Watchers+3

Early Traction+0

Devto Reactions+0

Community Contribs+2

Scope

Commits+7

Languages+8

Subsystems+5

Bloat Penalty+0

Completeness+7

Contributors+6

Authored Files+12

Readme Code Match+3

Architecture Depth+7

Implementation Depth+8

Evidence

Commits

Contributors

Files

Active weeks

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

Forks

License

AGPL-3.0