Skip to main content
AIDiveForge AIDiveForge

Share This Tool

Compare This Tool
📋 Embed this tool on your site

Copy this code to embed a compact tool card:

Call Me

Summary

Call Me runs large language models as voice agents that place and handle phone calls on behalf of users.

The tool connects an LLM to telephony infrastructure so the model can speak, listen, and respond during live calls. It sits in the narrow niche of replacing manual dialing and scripted conversations with automated yet flexible dialogue. No public pricing numbers appear on the listing or linked pages. The largest drawback is the complete absence of performance data or usage examples, leaving callers without evidence of how well the system manages accents, interruptions, or unexpected replies.

Bottom line: *Choose Call Me only after confirming costs and testing call quality on low-stakes tasks.*

Community Benchmarks Community

No community benchmarks yet. Be the first to share a real-world data point.

Community Reviews

No reviews yet. Be the first to share your experience.

Discussion Community

No discussion yet. Sign in to start the conversation.

Spotted incorrect or missing data? Join our community of contributors.

Sign Up to Contribute

Community Notes & Tips Community

Be the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.

Hours Saved & ROI Stories Community

Be the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."

Most agent sessions assume you are watching. The moment a Claude Code job runs unattended — overnight, during a meeting, across a slow API pipeline — there is no channel back to you when the agent hits a decision point. Call Me closes that gap by pairing a Claude Code plugin with a free iOS app, giving agents a real phone number they can call or text. Setup is two slash commands inside Claude Code and a QR-code scan on your phone. After a fresh session restart, the inbound monitor comes online and the agent can initiate calls, speak prompts, and return your voice response as text into the running session.

The differentiating detail is directionality: the agent initiates the contact, not you. Most notification tools push one-way updates. Here the agent can block on your answer — ask a branching question mid-task, get a spoken reply, and continue. That loop is what makes it useful for genuinely long-running tasks rather than just status pings.

The fit is narrow and honest about it. The repo is built for Claude Code users on iPhone. The plugin installs from the Claude Code marketplace; the iOS app is free on the App Store. There is no REST API, no webhook surface, and no documented path for agents running in other frameworks. The inbound monitor that enables the agent-to-human call loop only initializes in a fresh Claude Code session — teams that skip the restart after install lose that capability without an obvious error.

On the technical side, the repo ships both a Claude Code plugin directory and a skill directory. A CLI companion (`callme qr`, `callme setup`) handles first-run configuration from the terminal. The project is open source under the radres org, so the agent-side integration logic is inspectable and forkable — relevant if you want to adapt the skill layer for a different workflow.