Thousands of Actors to automate your business, get real-time web data, and equip your AI agents. ➡️ https://xitter.cf/t.co/rySkh8pPkM • https://xitter.cf/t.co/PiUfLvrm0u • https://xitter.cf/t.co/x4G1meKqfK
The Interwebz
Joined September 2015
- Tweets5.2K
- Following305
- Followers19.9K
- Likes4.6K
Pinned Tweet
Training makes a great AI agent. Purpose-built tools make the mission a success.
🧰 Equip your AI on Apify → apify.it/X200v4
Apify retweeted
Here’s how to give your AI agent access to data from almost any website.
With Apify MCP, Claude can use 80,000+ tools to pull data from Google Maps, Instagram, LinkedIn, and more.
Watch here: youtu.be/c2QukPONnfE
Get started with Apify and use code TIM25 for $25 in free credits: apify.it/4xGIvHG
Ouch. Someone gave him a toy crossbow. Don't do that to your agents.
🧰 Equip your AI on Apify → apify.it/X200v2
Four extra form fields made every lead 34% more expensive in the Marketo test @richardrx breaks down.
The fix: ask for less and enrich the rest. His team pulled 10,000+ leads in Brazil with Apify for under $25 👇
Cada pergunta a mais no teu formulário tem preço, a Marketo no case que apresento abaixo, calculou isso em dólar.
Eles fizeram um teste multivariado, com 3 variações:
(1) 5 campos (nome, sobrenome, e-mail, cargo, empresa) = 13,4% de conversão, US$31,24 por lead
(2) 7 campos (+ nº de funcionários e setor)
= 12%, US$34,94 por lead
(3) 9 campos (+ telefone e qual CRM usa)
= 10%, US$41,90 por lead
4 perguntas a mais e cada lead ficou 34% mais caro, com o clique custando exatamente o mesmo.
A conclusão do VP de marketing deles foi de pedir menos e enriquecer os dados depois, citando o próprio ele escreveu
"Por que perguntar, derrubar a conversão e ainda receber dado errado, se dá pra comprar mais barato?"
No Brasil sai mais barato ainda, porque é de graça, pede o CNPJ e a base aberta da Receita te devolve razão social, porte e atividade principal.
O formulário geralmente pergunta "tamanho da empresa" e "setor" num dropdown está cobrando do lead (e comendo conversão) numa informação que o governo já publica.
Aqui eu passei a investir em créditos usando Apify (dica do @jeffecom aliás) e levantamos simplesmente mais de dez mil leads no Brasil por menos de 25 usd.
Transcript data from Reels. Business contacts from search. Account history behind any profile. Actor Spotlight, Instagram edition ↓
Apify retweeted
Blank connection requests on LinkedIn: 12% acceptance
Connection requests to active profiles: 20% acceptance
Simply:
1. Use actor from Apify to identify active profiles
2. Score list based on activity
3. Put active profiles at the top of your list
Easy.
How many tokens does your agent burn just reading its tool list?
@DivyanshT91162 walks through mcpc, Apify’s CLI that connects to Apify MCP through one Bash tool, finds tools on demand, and keeps credentials protected.
One install, and you’re in 👇
Your AI agent might be wasting thousands of tokens just figuring out which tools it can use.
Apify built a CLI to change how AI agents interact with MCP.
Meet "mcpc" — a universal MCP client that lets an AI agent access MCP servers through a single Bash tool instead of wiring up dozens of separate tool functions.
Here's why this is interesting:
→ Dynamic tool discovery: Find the tools you need instead of loading everything into context upfront.
→ Persistent sessions: Keep multiple MCP server connections alive in parallel.
→ Full MCP support: Access tools, resources, prompts, tasks, and skills through shell commands.
→ Security built in: OAuth 2.1, OS keychain credential storage, and a proxy to protect credentials from AI-generated code.
→ Composable workflows: Pipe JSON output into "jq", "xargs", and other shell utilities to build repeatable automations.
Get started:
"npm install -g @apify/mcpc"
Then connect to an MCP server:
"mcpc connect mcp.apify.com @apify"
The bigger idea?
Stop flooding your AI agent with tools. Let it discover the right ones when it needs them.
MCP provides the standard. The CLI gives agents a familiar interface to use it.
REPOO👇
AI engineering has moved from novelty to standard practice - the industry is figuring out how to make agents dependable. And dependable agents need proper tools. Three days at AIE NYC well spent.
Thanks, @aiDotEngineer team 🙌
An Actor is not a scraper. Get that wrong, and nothing about Apify makes sense.
It’s a cloud program - input in, data out, no server to own. Hand one to an AI agent, and get the job done.
Equip your AI agents with thousands of tools on Apify ↓
apify.it/4dNfinx
Every agent needs proper tools. @TechWithTimm connects @claudeai to Apify MCP, pulls live data from five platforms for $2.71 total, then builds a skill that finds and pitches new clients.
What would you have your agent find first? 🤔
Here’s how to give your AI agent access to data from almost any website.
We’ll connect Apify through MCP, then use it to pull data from Google Maps, Instagram, TikTok, LinkedIn, and YouTube.
At the end, I’ll combine those tools into a skill that finds and scores potential clients.
Timestamps:
00:00 | Overview
00:27 | The Problem
01:36 | What Is Apify?
03:16 | Apify MCP
03:59 | Connecting Apify to Claude
05:42 | Demo 1: Google Maps
07:34 | Demo 2: Instagram
08:32 | Demo 3: TikTok
09:21 | Demo 4: LinkedIn
10:18 | Demo 5: YouTube
11:32 | Combining Everything into a Skill
Seeing Apify help ideas come to life is the best part of any event. Thanks for the kind words, @NadineCodes! 🙌
