Call Me
Summary
Call Me is a closed-source agent framework that adds telephony to LLMs for outbound and inbound voice calls.
The system wires language models to phone networks so agents can dial numbers, speak with people, and complete tasks like reminders or basic support without separate voice infrastructure. It sits in the gap between chat-only LLM agents and production voice systems that need reliable call handling. Pricing is not publicly disclosed. The largest drawback is the absence of documented benchmarks, adoption numbers, or cost details, making evaluation dependent on direct outreach rather than public evidence.
Bottom line: *Choose Call Me only when voice telephony must be added to an LLM agent; skip it for text-only workflows or when cost clarity is required.*
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.
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
Long-running agents that pause for human input usually do so badly — a blocking prompt in a terminal, a webhook nobody monitors, a Slack message buried under standup noise. Call Me gives Claude Code agents a direct line to an iPhone: the agent calls the number, speaks a question using voice synthesis, the user answers out loud, and the response comes back as text the agent can act on. Text-only notifications work too, which means a cron-triggered script can send an SMS and wait for a reply before proceeding. The core workflow is three setup steps: install the free iOS app, copy the personal Call Me number it displays, then send two slash commands to Claude Code — the docs state Claude handles the rest of the configuration itself after a single instruction message.
The differentiating capability is the inbound reply loop. Most notification tools are one-way: the agent fires a message and moves on. Call Me lets the agent block until you respond, which means decisions that genuinely require a human — approve this deployment, confirm this purchase, pick option A or B — stay in the agent’s execution flow rather than requiring a separate re-trigger mechanism. The vendor describes this as enabling agents to reach the user hours after the original task started, which covers the class of background jobs that cannot know in advance when they will need input.
This fits tightly into a solo-developer or small-team context where the agent runner and the human decision-maker are the same iOS user. It breaks at two boundaries: platform (iPhone only, by design) and audience (it gives agents a way to reach the developer, not the developer’s end users). Teams building agents that need to loop in a product manager, a client, or anyone without an iPhone will need a different confirmation channel. Because there is no hosted API or commercial tier, teams that need auditability, access controls, or multi-user routing will find themselves building that layer themselves.