Call Me
Summary
Call Me runs large language models as autonomous phone agents for outbound voice tasks.
The service connects an LLM directly to telephony so the model can place calls, speak, listen, and respond without a human on the line. It sits in the narrow niche of agent frameworks that move language models from text to real-time audio channels. Its stated differentiator is the direct model-to-phone link rather than layered wrappers. No public pricing figure is listed. The clearest drawback is the closed-source restriction that prevents inspection or self-hosting of the underlying agent logic.
Bottom line: *Choose it to prototype LLM voice calls; skip it when cost transparency or code access 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
Most agents that need a human answer mid-run either block indefinitely or dump a log somewhere nobody reads. Call Me solves that by letting agents dial your real iPhone, speak a question, and receive your spoken answer as transcribed text — or send a push notification or SMS if a voice call is too much. The agent-side integrations live in this repo: a Claude Code plugin installed via two slash commands, and a skill file for other agent frameworks. The iOS app is free and generates a personal 10-digit Call Me number; that number is the only configuration an agent needs.
The differentiating detail is the call-and-transcribe loop. An agent can ask a spoken question, wait for you to speak an answer, and receive that answer as text it can act on — not a button tap, not a form field. For agents running unattended CLI scripts or cron jobs where you want sign-off before a destructive action, this replaces the awkward ‘check your email’ placeholder with something that actually interrupts you.
The tool fits tightly scoped solo or small-team workflows: a Claude Code session where you want phone reachability, a cron job that needs your approval before it writes to production, or an agent that should notify you when a long task finishes. It breaks when the approval chain is more than one person, when reviewers are on Android, or when the volume of interruptions exceeds what a phone call pattern can absorb. There is no web interface, no team inbox, and no audit log of past calls described in the available documentation.
The vendor states the CLI includes a `callme qr` command that prints a scannable QR code for the App Store link, and a `callme setup` command that walks through configuration steps. The Claude Code integration stores your Call Me number after a one-time setup conversation and confirms it with a test text.