Lists (1)
Stars
Taste-Skill (High-Agency Frontend) - gives your AI good taste. stops the AI from generating boring, generic, "slop"
Give your AI agent access to your live Chrome session — works out of the box, connects to tabs you already have open
Growing Expertise for Coding Agents — structured expertise files that accumulate over time, live in git, work with any agent
pi extension to empower a curl request with coding agent capabilities (slopcoded)
A simple mobile-focused chat app to talk to an agent via the ACP protocol
These are commands I use with agents, mostly Claude
Awesome list of add-ons, hooks, tools, skills, and resources for the pi coding agent (pi-mono).
Skills for pi coding agent (compatible with Claude Code and Codex CLI)
An AI agent that stays running, remembers across sessions, and checks in on its own. macOS, Linux, Android. Built on Pi.
AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods
AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
Unified TUI and CLI to index and search your local coding agent session history across 11+ providers (Codex, Claude, Gemini, Cursor, Aider, etc.)
A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.
Transcripts from Lenny's Podcast
An agentic skills framework & software development methodology that works.
omo; the best agent harness - previously oh-my-opencode
A configuration framework that enhances Claude Code with specialized commands, cognitive personas, and development methodologies.
mini cli search engine for your docs, knowledge bases, meeting notes, whatever. Tracking current sota approaches while being all local
Prompts, workflows and more for agentic engineering
Official inference framework for 1-bit LLMs
Orchestration framework for running 20+ Claude Code agents in parallel: automated bug fixing, best-practices sweeps, lock-based coordination, and real-time tmux monitoring
Detailed Next.js best-practice guides designed for humans and LLM coding assistants to collaboratively build robust applications