Call Me
Summary
Call Me runs large language models as voice agents that make and receive phone calls.
Users configure LLM agents to conduct complete telephone conversations in place of human callers or rigid IVR systems. The tool targets the gap between text-based chatbots and real-world telephony needs where natural back-and-forth dialogue is required. No pricing information is published, so expected cost cannot be calculated in advance. Its closed-source status and lack of public benchmarks or model details form the clearest limitation on evaluating actual performance.
Bottom line: *Use it for quick LLM voice experiments; skip it when call reliability or cost transparency matters.*
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Call Me
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Agents that make autonomous decisions hit a class of problems where the right answer is not in the context — it’s in your head, and they need to ask. Call Me is an open-source integration that lets an AI agent dial your iPhone, speak a question via text-to-speech, and receive your spoken answer back as transcribed text, or alternatively send and receive an SMS. The core workflow: install the iOS app (free on the App Store), copy your personal Call Me number from it, install the Claude Code plugin with two slash commands, tell Claude to run setup, and the agent has a persistent way to reach you. Confirmation arrives as a text, and the number is stored so subsequent calls need no reconfiguration.
The differentiating feature is the call-and-transcribe loop. Most notification tools push a message and stop — you see it when you see it. Call Me rings the phone, waits for a spoken response, and hands that transcription back to the waiting agent as structured input. That loop is what makes mid-task branching on human input possible rather than theoretical.
The tool fits tightly into Claude Code workflows and any agent framework that can invoke a bash skill or CLI command. The repo ships a skill file specifically for that second case, so cron-driven scripts and non-Claude agents that support shell execution can use the same call/text mechanism. Where it breaks: the receiving side is iOS only — there is no Android app, no web fallback, and no API endpoint for teams wanting to route approvals to a queue or share them across multiple people. A solo developer who owns an iPhone and runs agents locally gets the full value. A team that needs shared approval routing, or whose members carry Android phones, will need a different architecture.