KonH/SmartReader

View on GitHub by @KonH

Personalized news aggregator that reads RSS and Telegram, scores and summarizes articles with AI, and learns your interests from feedback

What's novel

Personalized news aggregator that reads RSS and Telegram, scores and summarizes articles with AI, and learns your interests from feedback

Code Analysis

19 files read · 4 rounds

A personalized content aggregator that reads RSS/Telegram feeds, scores articles using a two-tier system (fast keyword pass + LLM refinement), merges related articles, and learns user preferences from upvote/downvote feedback to improve future recommendations.

Strengths

Clean architecture with abstract interfaces and swappable implementations, practical two-tier scoring that balances cost and quality, and a well-designed feedback loop that continuously improves the interest model. The code is readable, well-organized, and the callback-based async pattern is pragmatic for a CLI/bot application.

Weaknesses

The SourceReader.read_sources method has a race condition where it returns empty results because it calls the callback before the async readers complete. Test coverage is good for core logic but missing for LLM-dependent code paths and the Telegram UI.

Score Breakdown

Innovation

6 (25%)

Craft

71 (35%)

Traction

8 (15%)

Scope

82 (25%)

Signal breakdown

Innovation

Not Fork+1

Code Novelty+1

Concept Novelty+2

Craft

Ci+0

Tests+8

Polish+1

Releases+0

Has License+5

Code Quality+23

Readme Quality+15

Recent Activity+7

Structure Quality+5

Commit Consistency+2

Has Dependency Mgmt+5

Traction

Forks+0

Stars+6

Hn Points+0

Watchers+0

Early Traction+0

Devto Reactions+0

Community Contribs+2

Scope

Commits+8

Languages+5

Subsystems+10

Bloat Penalty+0

Completeness+7

Contributors+6

Authored Files+15

Readme Code Match+3

Architecture Depth+7

Implementation Depth+8

Evidence

Commits

Contributors

Files

132

Active weeks

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

Forks

License

MIT