Claude CodeRepo★25k
12-factor-agents
by humanlayer ✓
Summary
Practical principles for building reliable, production-ready LLM agents.
Content
Twelve principles to help an LLM agent survive production, written by people who have already been burned: explicit control over the prompt, the context and the flow, instead of leaving it to the framework. The most useful part for an agency is the diagnosis, because it explains why an agent that works in the demo falls over with real clients. Watch the date: the repository has had no changes since September 2025, so read it as a conceptual framework and not as a library reference.
View official pageHow to use
Follow the README of the original repo (link below). This directory only indexes and gives credit; the code lives at the source.
Recommended
See more →
Agent Skills (official repo)
Anthropic's official repo of public Agent Skills: reusable skills that teach your agent tasks.
#claude-code#skills#oficial
RepoClaude Code★164k

agents (multi-harness subagents)
A collection and marketplace of subagents and plugins for Claude Code, Codex and other agents.
#claude-code#codex#subagents
RepoClaude Code★38k

Agno
A lightweight, fast framework for building agents with memory, tools and RAG.
#agentes#framework#rag
RepoGeneral★41k
More resources like this
Get the newest prompts and tools every week.