call4me blog

Read the call4me blog (https://call4.me/blog, every post as markdown in https://call4.me/llms.txt) and help me get a phone chore done with call4me.

If you don't have the call4me tools (call4me_*) yet, add the call4me MCP server first and sign me in:
   - Claude Code:  claude mcp add --scope user --transport http call4me https://call4.me/mcp  (then I run /mcp, pick call4me, and choose authenticate)
   - Codex:        codex mcp add call4me --url https://call4.me/mcp  then  codex mcp login call4me
   - Claude Desktop, claude.ai, or ChatGPT: add a custom connector with the URL https://call4.me/mcp and sign in when it asks.
   - Muse: build a custom integration to the call4me remote MCP server at https://call4.me/mcp (streamable HTTP) with the official MCP SDK. Ask me for my call4me API key (it's at https://call4.me/account) through your secure credential flow, never in chat, store it in your Secure Credentials Store, and send it as Authorization: Bearer <key>. List the tools, test call4me_get_balance, and save it as a reusable skill.
   - Grok Bot: add a custom MCP server called call4me (remote HTTPS, no headers, no auth) at my personal server URL, which I copy from https://call4.me/account. You add it from this chat; no separate approval is needed. If you offer sign-in instead, use https://call4.me/mcp and I sign in.
   - grok.com: you can't add connectors from the chat, so I do it: grok.com/connectors, New Connector, Custom, with the URL https://call4.me/mcp, then I sign in.
   - If you can't sign me in, I copy my personal server URL from https://call4.me/account and you use that instead.
   I sign in with Google or X.

Ask me what I need called for. If a post covers it, use its tips. Then call call4me_get_requirements for the category, ask me in one message for anything missing, place the call with call4me_place_call, follow it with call4me_get_call until it finishes, and tell me the result in one or two lines.

notes on AI agents that make phone calls for you.

Superintelligence Calling Index: compare personal superintelligence phone calling tools, hear the evidence and download the dataset.

latest toparchive RSS

44 posts matching "return policy". clear

How to get your AI assistant through a phone tree: Vapi, Bland and call4me calls

A phone menu, a successful keypad press, and a completed task are different results. Here is how to configure, diagnose, and test each step.

OCT 4 / Nick Khami

How to make phone calls from Codex with MCP

Give Codex a calling tool, keep your credentials out of the prompt, and bring phone answers back into the research it is already doing.

OCT 3 / Nick Khami

Phone calling MCP servers for Claude Code and Codex: which setup fits your agent?

Compare call4me, Vapi Agent Phone, Bland, ClawCall, Cocall, and Patter by account setup, calling workflow, and the evidence your agent gets back.

OCT 3 / Nick Khami

Cascaded voice stack vs GPT Live for phone calling agents

What changes when a phone agent hears and speaks directly, delegates work to a separate reasoner, and has to get through a real phone menu.

OCT 3 / Nick Khami

Fabletics customer service: how to reach a person, skip the VIP charge or cancel

A recorded call about the $69.95 VIP charge, skipping a month and what cancellation questions required an account.

OCT 2 / Nick Khami

How to add a phone calling MCP server to Grok and Grok Bot

Set up custom tools on grok.com, in Grok Bot or through the xAI API. Then hear our actual Grok Bot restaurant call, including the partial answer and the menu problem.

OCT 1 / Nick Khami

Muse MCP: connect Meta Muse to phone calling tools

Connect Meta Muse to a phone calling service through a custom connector. Includes our tested setup, successful reuse in a fresh conversation and a recorded library inquiry.

OCT 1 / Nick Khami

How to use an American Airlines flight credit by phone

A call4me user's aa.com checkout wouldn't take their flight credit, so their AI phone assistant called American Airlines reservations twice to apply it to a held trip. Here's who has to be on the line, why the credit was worth less than the full ticket, both recordings, and what to have ready.

OCT 1 / Nick Khami

recommendations

Anthropic engineering - how Claude Code and agent tools are built, from the people building them.
Simon Willison's weblog - daily, careful notes on what LLM tooling actually does.
Latent Space - the AI engineer newsletter and podcast.
The Cloudflare blog - agent readiness, MCP, and the plumbing of the agentic web.
Stripe blog - machine payments and how agents pay for things.