Everything AI agents, in one place.
Knowledge, tools, playbooks, templates, and automation loops — curated so you can start building today.
Start here
Freshly picked
What is Agentic AI?
Agentic AI describes systems where language models plan, use tools, and act toward a goal, looping until the task is done. Here is what changes when you move from chatbots to agentic workflows.
Deep-research agent in a weekend
Assemble a multi-step web researcher with citations.
Customer support copilot
Ground a support agent in your docs and past tickets.
Personalized outbound at scale
Prospect, enrich, and draft — with a human in the loop.
Self-hosted code review bot
Ship a PR reviewer that comments like a senior engineer.
Meeting notes to action items
Turn transcripts into tasks in your PM tool.
More to explore
Researcher system prompt
A battle-tested system prompt for web research agents.
Support agent system prompt
Grounded, polite, escalates when unsure.
Search tool JSON schema
Ready-to-paste tool schema for OpenAI/Anthropic function calling.
LangGraph supervisor template
Python starter for a supervisor + workers graph.
RAG pipeline scaffold
Chunk → embed → retrieve → answer, with citations.
Golden set spreadsheet
A schema for evals: input, expected, judge notes.
New lead → enrich → notify
When a form is submitted, enrich the lead and post to Slack.
New ticket → classify → route
Auto-classify Zendesk tickets and route to the right queue.
Daily → summarize repos → post
Every morning, summarize yesterday's commits into Slack.
Blog published → repurpose → schedule
Turn a new blog post into a thread + LinkedIn post.
Invoice overdue → draft email → send
Politely nudge overdue invoices with a human review step.
What are AI agents?
A working definition, the observe-act-observe loop, and when to reach for an agent instead of a workflow.
Build an inbox triage agent
Classify, label, and draft replies to every email you get.
Tool use, function calling & MCP
Function calling, JSON schemas, and MCP — the plumbing that lets an LLM reach the outside world.
Memory in agents
Short-term context, long-term retrieval, and scratchpads — the three memory layers every agent needs.
Multi-agent orchestration
Supervisor, hierarchical, swarm, and role-based topologies — plus when a single agent is still the right call.
Evaluating agents in production
Golden sets, LLM-as-judge, trace-based debugging, and the tooling that makes it repeatable.
Securing your agents
Prompt injection, excessive agency, and the least-privilege patterns from the OWASP LLM Top 10.
Prompt engineering for agents
What still moves the needle in 2026 — structure, examples, and constraints that make agents behave.
LangGraph
Open-source graph-based orchestration for stateful, multi-actor LLM applications. Made by LangChain.
CrewAI
Role-based multi-agent framework in Python — define crews with goals, tasks, and tools.
OpenAI Agents SDK
Official OpenAI SDK for building agents with handoffs, guardrails, and tracing built in.
Claude Code
Anthropic's coding agent that lives in your terminal and edits your repo with your permission.
Cursor
AI-first code editor with in-repo chat, agent mode, and MCP support.
n8n
Open-source workflow automation with 400+ integrations and a native AI/agent node.
Make
Visual automation platform with an AI module for Anthropic, OpenAI, and more.
Zapier Agents
No-code agent builder from Zapier with access to 7,000+ app integrations.
Langfuse
Open-source LLM engineering platform — tracing, prompts, evals, and datasets.
mem0
Memory layer for AI agents — persist user facts across sessions and models.
Agents vs. workflows
When a predefined workflow beats a dynamic agent — with the patterns from Anthropic's canonical guide.