Yell Repo
A curated collection of the most interesting GitHub repositories — from AI agents and Claude Code workflows to local LLMs and research tools. Each project is explained as a story: what it does, why it matters, and when you should use it.
One Source, Two Surfaces: Anthropic's Financial Services Toolkit
10 Claude agents for investment banking, equity research, and fund admin, deployable as Cowork plugins or headless Managed Agents from a single YAML manifest.
OpenGame: The Agent That Runs the Game to See If It Works
OpenGame is a CLI agentic framework from CUHK MMLab that scaffolds, builds, and verifies browser games end-to-end from a plain-English prompt.
Pixelle-Video: One Topic, a ComfyUI Assembly Line
Streamlit app that turns a topic or script into short video: LLM copy, ComfyUI or RunningHub media, TTS, BGM, HTML templates, ffmpeg out to output/.
Maigret: The OSINT Detective That Finds Every Account
OSINT tool that checks 3,000+ platforms for a username, recursively hunts linked accounts, and generates full HTML/PDF dossiers; no API keys required.
Four Principles to Stop LLMs Overengineering Your Code
A tiny CLAUDE.md and optional plugin that makes Claude Code ask, test, and edit surgically.
Android Reverse Engineering Skill: Unpacking APKs Like a Pro
A Claude Code skill that decompiles Android apps and extracts their hidden APIs - no source code required. For security pros, hackers, and the curious.
Blitz: Autopilot for App Store Submission
Native macOS App Store Connect client that automates submissions, screenshots, IAPs and agent-driven workflows via MCP.
Kronos: A Foundation Model for the Language of Financial Markets
Kronos tokenizes OHLCV K-line data and provides pre-trained autoregressive models and finetuning tools for forecasting.
The Job Search System That Evaluated 740 Offers
Career-ops turns Claude Code into a 14-mode job search agent: structured A-F evaluations, ATS-optimized PDFs, batch processing, and a Go TUI dashboard. MIT.
Context Hub: The Curated Doc Layer Every Coding Agent Needs
Context Hub (`chub`) gives AI coding agents current, versioned API docs on demand -- so they code from facts, not 18-month-old training weights.
MarkdyScript: The Mermaid of Motion
MarkdyScript is a declarative text DSL that turns 30 lines of plain code into Web Animations API scenes, git-diffable, LLM-friendly, and zero GSAP.
MarkItDown: Everything Becomes Markdown
Microsoft's Python tool converts PDFs, Office docs, audio, and YouTube URLs to clean Markdown for LLM pipelines and RAG systems.
taste-skill: The Bias Correction File for AI Frontend
taste-skill corrects the font, layout, and motion biases that push LLMs toward generic UI. A portable SKILL.md you commit once per project.
GitNexus: The Repo Map That Finally Helps Agents Navigate
GitNexus builds a local code graph for MCP tools, browser exploration, and impact checks when text search stops explaining the repo.
Claude Code's Missing Manual: Stop Typing and Start Orchestrating
A copy-paste survival guide that turns Claude from a basic chatbox into an automated team member.
AI Scientist-v2: The Robot That Writes Its Own Papers
SakanaAI's AI Scientist-v2 autonomously generates ML research hypotheses, runs experiments via agentic tree search, and writes peer-reviewed papers. No…
The Agent is the Model. The Code is Just a Vehicle.
A deep dive into learn-claude-code and the truth about AI agents: most frameworks are over-engineered Rube Goldberg machines.
The Conductor and the Machine: Orchestrating Claude Code
Claude Code is a beast. But to tame it, you need more than prompts. You need an orchestration framework.
Claude Skills: The Surgical Precision of Agentic Experts
A library of 200+ modular expert instruction sets and zero-dependency Python tools for AI coding agents.
VibeVoice: The Hour-Long Storyteller
AI that doesn't forget who is talking even after an hour. VibeVoice handles long-form speech without the usual 'memory loss'.
The Professional Kitchen for Your AI Agent
Everything Claude Code isn't just a config pack; it's a performance system that turns a general-purpose AI into a precision instrument.
Hermes Agent: The Desk That Remembers
Most AI agents reset every session. Hermes Agent builds a persistent workbench that learns your stack, grows its own skills, and follows you across 7 platforms.