
Heard is a macOS voice layer for your agentic workflows. It connects to Claude Code, Codex, and Cursor and turns their output into intelligent summaries you can hear. Full commentary when your eyes are elsewhere, or silence until something errors or needs a decision. With agents run in parallel, Heard summarizes at the project level, so you hear where the work stands, not five terminals talking over each other. Pair your phone and it comes with you. Open source, free for personal use.
Loading comments…
Project Info
Product Keywords
Heard is a macOS voice layer designed to bring ambient intelligence to your terminal. It connects directly to coding agents like Claude Code, Codex, and Cursor, monitors their event streams, and transforms what matters into concise, natural spoken summaries. Instead of watching a terminal window or wading through raw logs, you get intelligent narration — full commentary when your eyes are elsewhere, silence until something errors or needs a decision, and project-level summaries when multiple agents run in parallel. Heard is open source and free for personal use.
Heard offers three distinct modes switchable from the menu bar. Co-pilot provides short hooks and signposts while you work — enough to keep pace without narrating every keystroke. Companion delivers fuller briefings for times your eyes are off the screen. Focus stays quiet until something needs you: an approval, a blocker, a failure, or a decision waiting to be made.
Pair your phone with a one-time scannable code, and live narration follows you anywhere — even with the screen locked. You can listen, approve, and reply to your agents by holding or tapping to talk. On-Mac speech recognition keeps developer vocabulary intact, and scoped pairing with automatic disarm ensures safety.
Heard follows parallel agent sessions simultaneously, voices the most important signal, and rolls the rest into compact background digests. Distinct voices make each agent recognizable by ear, so you hear where the work stands — not a chaotic stream of overlapping output.
Choose from four built-in personas or write your own in a Markdown file. You can also pick the speaking voice, speed, and listening mode that match your temperament for the day.
Heard turns agent activity into clear spoken updates, so you can step away without falling behind.
Most AI coding tools focus only on input — how you talk to the agent. Heard solves the neglected output side: it listens with judgment, filters noise, and delivers context-aware summaries that actually help you stay productive away from the screen. Its ability to handle multiple agents in parallel, pair with your phone, and auto-pause during calls makes it feel less like a notification system and more like a thoughtful co-worker.
You use Claude Code, Codex, or Cursor and find yourself tied to the terminal waiting for agent output. Or if you run several AI coding sessions simultaneously and want a single, coherent audio feed instead of juggling terminal tabs. Heard is especially useful if you value stepping away from your desk without losing context — whether that means heading outside, checking in from your phone, or simply keeping your focus on the code while your ears stay informed.
Other tools you might consider
You're running more coding agents than ever, but you can't keep up with them. That's where AgentPeek comes in. It pulls every session up into your Mac notch, live. Glance up, approve a prompt, watch token usage and manage the entire flow without pausing your YouTube video. All local, all yours.
Navigara connects AI coding performance directly to your engineering roadmap. Analyzing code like a senior engineer to prove real capacity gains, Navigara tracks exact costs per roadmap item, isolates off-roadmap waste, and identifies maintenance burn. Cut spend further by automatically routing routine CRUD tasks to low-cost models without sacrificing quality. Connects in minutes via Git history, JIRA/Linear, and any AI coding license for spend.
GitHits gives coding agents access to the open-source code your app depends on. Get real implementation examples, dependency source navigation, package inspection and documentation. Agents can grep and read your codebase. They can't grep and read the open-source code your app depends on. That's where they start guessing, retrying, and looping. GitHits builds a version-aware index on demand. Agents can search, navigate, and inspect the code behind their dependencies. CLI: npx githits@latest init
The Supercut MCP gives your AI/coding assistants permission-aware access to recordings, including semantic search, transcripts, frames, comments, reactions, and more.
Maker
indie_inkwell
Alternatives
Loading comments…