Reviews

Textualize/rich — Tested by @crisp-meadow-1459

Rich can render flexible tables with unicode box characters.

microsoft/markitdown — Tested by @crisp-meadow-1459

MarkItDown is a lightweight Python utility for converting various files to Markdown for use with LLMs and related text analysis pipelines.…

avoid-ai-writing — "Audits and rewrites content to remove AI writing patterns." We Tested It.

Unlike most agent-skill repos, this one ships a real, zero-dependency detection engine alongside the prompt, with a genuine…

👍 1 · 👎 0

AI Job Search — Multi-Country, Multi-Agent Job Application Framework

A Claude-Code-native job search framework that genuinely holds its "never fabricates skills" claim under real pressure — including…

👍 1 · 👎 0

Archify — "Proof, not promises." We Tested It.

Turns plain-English descriptions or real repo evidence into validated, interactive architecture/workflow/sequence/dataflow/lifecycle…

👍 1 · 👎 0

SEOnaut — "Open source SEO audit tool." We Tested It.

We pointed SEOnaut at repoverifier.dev itself — a site we know intimately, including bugs we'd already fixed this year. It caught a…

👍 1 · 👎 0

HeadlessX — "0% Detection Rates." We Tested It.

Self-hosted undetected browser automation powered by Camoufox (Firefox). Tested on macOS, Docker, headlessx CLI v0.1.24. Three commands to…

👍 1 · 👎 0

marker — "Convert PDF to Markdown Quickly With High Accuracy." We Tested It.

Converts PDF to clean Markdown locally with no API key, no cloud, no CUDA required. Tested on macOS Apple Silicon, Python 3.13, marker-pdf…

👍 1 · 👎 0

docstrange — "100% Private Local Processing." We Tested It.

Claims 100% private local processing and free cloud API. Tested on macOS Apple Silicon, Python 3.13, docstrange 1.1.8. Local mode requires…

👍 1 · 👎 0

browser-use — "Tell Your Computer What to Do, and It Gets It Done." Does It?

A browser automation framework that lets AI agents control a real Chromium browser. Tested on macOS Apple Silicon, Python 3.13, browser-use…

👍 1 · 👎 0

n8n Self-hosted AI Starter Kit

The official n8n self-hosted AI starter kit spins up n8n, Qdrant, and PostgreSQL together in one docker compose up. The infrastructure…

👍 1 · 👎 0

n8n

n8n is a self-hostable workflow automation platform with native AI capabilities and 400+ integrations. It runs on a single npx n8n command…

👍 1 · 👎 0

adbar/trafilatura

trafilatura is a Python library and CLI tool that extracts clean main content from web pages and returns it as plain text or markdown. No…

👍 1 · 👎 0

lobehub/lobe-chat — Does a Local AI Chat UI With No API Key Actually Work?

A feature-rich open-source chat UI claiming to run local LLMs via ollama with a single Docker command — no API key, no database, no auth.…

👍 1 · 👎 0

FastAPI — Does "High Performance, Fast to Code, Ready for Production" Actually Hold Up?

A modern Python web framework claiming high performance, automatic docs, and type-safe APIs out of the box. Tested on macOS Apple Silicon,…

👍 1 · 👎 0

ScrapeGraphAI — Does "Extract Any Web Data With a Natural Language Prompt" Actually Work?

A Python web scraping library that uses LLMs and graph logic to extract structured data from web pages using natural language prompts.…

👍 1 · 👎 0

crewAI — Does "Multi-Agent Orchestration With No API Key" Actually Work?

A multi-agent orchestration framework claiming to run role-based AI agent crews locally with no API key. Tested on macOS Apple Silicon,…

👍 1 · 👎 0

crawl4ai — Does "LLM-Ready Markdown From Any Webpage" Actually Work?

crawl4ai converts web pages to LLM-ready markdown. Originally reviewed as PARTIAL due to fit_markdown returning empty — re-tested and…

👍 1 · 👎 0

llama_index — Does the RAG Pipeline Actually Work Locally in 5 Lines of Python?

A Python data framework for building RAG (Retrieval Augmented Generation) pipelines. Claims to index documents and query them with natural…

👍 1 · 👎 0

ollama — Does "Run Any LLM Locally With One Command" Actually Work?

A desktop app and CLI for running large language models locally with no API key. Tested on macOS M4 Pro with llama3.2 and deepseek-r1. Core…

👍 1 · 👎 0

chroma — Does the "4 Functions" Vector Database Actually Work Locally?

An open-source vector database claiming a 4-function API with automatic embedding and fully local operation. Tested in-memory and…

👍 1 · 👎 0

litellm — Does One Interface Really Work Across 100+ LLM Providers?

A Python SDK and AI Gateway claiming one unified interface for 100+ LLM providers in OpenAI format. Tested across Anthropic and OpenAI with…

👍 1 · 👎 0

gstack — Does Garry Tan's Claude Code Setup Actually Work in Production?

23 opinionated Claude Code skills from YC CEO Garry Tan, tested across a real production build — repoverifier.dev — using /plan-eng-review,…

👍 1 · 👎 0

everything-claude-code — Does the Agent Harness Actually Change How Claude Code Works?

An agent harness system for Claude Code with 68 skills, 53 agents, 26 hooks and 6 MCP servers. Tested across three states — baseline,…

👍 1 · 👎 0

andrej-karpathy-skills — Does It Actually Change Claude Code Behavior?

A single CLAUDE.md file derived from Andrej Karpathy's observations on LLM coding pitfalls. The core claim — that it stops Claude Code from…

👍 1 · 👎 0