Agent Runtime

Cloudflare Free App Stack

Intermediate

Deploy a free-tier app with Cloudflare Static Assets, Worker API, D1 database, token-protected admin checks, one-command deploy and production E2E verification.

Build time

20-40 minutes after Cloudflare credentials are ready

Stack

Cloudflare Workers + Workers Static Assets + Cloudflare D1 + Hono + GitHub template

A LangGraph agent that wakes on a PagerDuty page, correlates logs, deploys, and config history across services, forms a root-cause hypothesis, and posts a triage summary to Slack so the on-call engineer skips the manual correlation step.

Build time

A weekend for a scoped prototype on one service

Stack

LangChain | LangGraph + PagerDuty + Slack API + Amazon CloudWatch + Amazon Bedrock

Cloudflare AI Gateway Free LLM Fallback Chain

Advanced

Route coding agents through a Cloudflare-hosted OpenAI-compatible gateway backed by verified free or free-credit LLM providers, with BYOK keys, model aliases, dynamic fallback, and a no-key reserve route.

Build time

2-3 hours after Cloudflare AI Gateway permissions are ready

Stack

Cloudflare AI Gateway + NVIDIA NIM + Mistral AI + Google AI Studio + Cloudflare Workers AI + Kilo Code Gateway + Ollama Cloud

Vercel Agent AI Gateway Central Hub

Intermediate

Deploy a free Vercel-hosted OpenAI-compatible gateway that keeps provider keys in one place, exposes stable model aliases to all agents, and fails over across verified free-tier LLM routes.

Build time

60-90 minutes after provider keys are ready

Stack

Vercel Functions + NVIDIA NIM + Mistral AI + Cloudflare Workers AI + Kilo Code Gateway + Ollama Cloud

A CLI workflow that turns plain-English requests like 'terraform for a highly available EKS' into ready-to-review IaC using aiac, with provider backends for OpenAI, Amazon Bedrock, or a local Ollama model so prompts stay inside your environment.

Build time

30 minutes to first generated module

Stack

aiac + Terraform + Amazon Bedrock + Ollama

An LLM agent with read-only kubectl tools that answers 'which cluster is this workload in?' across dozens of clusters, turning a daily manual hunt across 30+ contexts into a single natural-language query, with results delivered to chat.

Build time

An afternoon for a single-task helper

Stack

LangChainGo + Kubernetes + kubectl

An ambient email assistant that triages your Gmail, drafts replies, schedules meetings, pauses for your approval on risky actions, and remembers your preferences over time using LangGraph.

Build time

A weekend to follow the full 4-notebook workshop

Stack

LangGraph + GitHub Models + Mistral AI (La Plateforme)

A LangGraph agent that takes a blog, GitHub, or YouTube URL, scrapes it with Firecrawl, drafts platform-native Twitter and LinkedIn posts with Claude, and pauses for human approval before scheduling through Arcade.

Build time

Half a day for the basic setup

Stack

LangChain | LangGraph + Firecrawl + Anthropic | Claude + Arcade + LangSmith + Slack + Supabase

A central supervisor agent coordinates specialized worker agents (a web researcher and a math/analysis agent), deciding which to invoke next and when the task is done, using langgraph-supervisor.

Build time

2-4 hours for a prototype

Stack

Qdrant Cloud Free Tier + Anthropic Startup Program + OpenRouter

A manager agent plans, delegates, and validates work across specialist agents (researcher, writer, editor) using CrewAI's hierarchical process, with tasks allocated at runtime rather than pre-assigned.

Build time

2-4 hours for a prototype

Stack

GroqCloud + OpenRouter + Qdrant Cloud Free Tier

A deterministic triage pipeline that wakes on each new Zendesk ticket, classifies it with an LLM, searches a Notion knowledge base, posts a structured alert to the right Slack channel, and writes tags, priority, and a draft reply back to Zendesk.

Build time

Half a day to a day for a prototype

Stack

Zendesk + Notion + Slack API + OpenRouter

Three handoff-driven agents (research, write, review) collaborate as a swarm via LlamaIndex AgentWorkflow, passing control to one another until a report is finished — with an orchestrator variant for tighter control.

Build time

2-4 hours for a prototype

Stack

Qdrant Cloud Free Tier + Anthropic Startup Program + OpenRouter

Parse complex product manuals, FAQs, and policies with LlamaParse, index them, and serve document agents that answer customer questions with citations, traceability, and confidence — deflecting repetitive tickets.

Build time

Half a day to a day for a prototype

Stack

LlamaIndex + LlamaParse + OpenAI Platform

A lead agent plans a research strategy, spawns parallel subagents with separate context windows to investigate different aspects simultaneously, then a citation agent attributes every claim — the orchestrator-worker architecture behind Claude Research.

Build time

1-2 days for a prototype

Stack

Anthropic Startup Program + Upstash | Redis + Qdrant Cloud Free Tier

Index a GitHub repository with LlamaIndex, store embeddings in a vector database, and answer natural-language questions about the code with citations back to the exact files.

Build time

1-2 hours for a prototype

Stack

LlamaIndex + LlamaHub (LlamaIndex tools) + Qdrant Cloud Free Tier + GitHub repository + Google AI Studio / Gemini API

Talk to your calendar in plain English: an MCP-connected agent that creates events, checks availability, manages invites, and imports events from images or PDFs, usable from Claude, Cursor, or any MCP host.

Build time

1-2 hours to connect and configure

Stack

GitHub Models + Anthropic Startup Program

A CrewAI crew where a researcher agent searches and scrapes the web and an analyst agent normalizes the findings into a structured, deduplicated leads table written to CSV or a database.

Build time

2-3 hours

Stack

CrewAI + Firecrawl + Serper + OpenAI

An agent-runtime workflow that upgrades a breaking dependency across a codebase: it localizes affected usages, applies fixes in a sandbox, runs the test suite, and opens a PR when green.

Build time

A day or more for a real codebase

Stack

OpenHands + E2B + GitHub repository + GitHub Actions + Anthropic Startup Program

An agent that takes a research question, fans out web searches, scrapes the most promising sources into clean markdown with Firecrawl, and synthesizes a cited report — an open-source alternative to OpenAI Deep Research.

Build time

Half a day for a prototype

Stack

Firecrawl + OpenAI Agents SDK + OpenAI + Anthropic Startup Program

A LangGraph state machine that decomposes a research question into sub-questions, finds sources with Exa semantic search, extracts and reranks passages, and writes an executive report with citations — auditable and private enough for finance or legal due diligence.

Build time

A day for a working pipeline

Stack

LangChain + Exa + Cohere + OpenAI

A CrewAI crew of specialist agents — an SEO analyst, a brand-positioning analyst, and a synthesis writer — that researches competitors in parallel and produces a coherent, fact-based strategic report instead of a single subjective prompt.

Build time

2-3 hours

Stack

CrewAI + Firecrawl + Serper + OpenAI

A research agent whose web-search tool is a self-hosted SearXNG metasearch engine instead of a metered commercial API — all search traffic stays local, no per-query billing, and you control which engines are queried, paired with Firecrawl for clean extraction.

Build time

Half a day including Docker setup

Stack

SearXNG + LangChain + Firecrawl + Ollama