by heardlabs
Open-source (Apache 2.0) voice tool for Claude Code and Codex on macOS that reads your coding agent's progress aloud. The voice runs locally with Kokoro, no account needed, or bring your own ElevenLabs or Speechify key. Hear a short spoken summary when your agent finishes, fails or needs approval.
# Add to your Claude Code skills
git clone https://github.com/heardlabs/heardGuides for using ai agents skills like heard.
Last scanned: 10/6/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-10-06T11:02:25.141Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how heard compares with popular alternatives.
heard is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by heardlabs. Open-source (Apache 2.0) voice tool for Claude Code and Codex on macOS that reads your coding agent's progress aloud. The voice runs locally with Kokoro, no account needed, or bring your own ElevenLabs or Speechify key. Hear a short spoken summary when your agent finishes, fails or needs approval. It has 884 GitHub stars.
Yes. heard passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/heardlabs/heard" and add it to your Claude Code skills directory (see the Installation section above).
heard is primarily written in Python. It is open-source under heardlabs on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh heard against similar tools.
No comments yet. Be the first to share your thoughts!
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM lists this project for discovery and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Automated security scan — not a certification.
A voice layer lets your coding agent talk to you, not only you to it. Claude Code's /voice covers the input half. Heard covers the output half: it follows the session and tells you out loud what changed, what failed and what needs a decision. The Heard app adds input too (on Power, press right ⌘ to talk to Parrot), so the whole loop can run by voice.
Can you use text-to-speech with Claude Code? Yes. Claude Code's built-in /voice command is voice input: you talk and it types your prompt. Heard adds the other direction. Run heard install claude-code and Heard adds hooks (Stop, PreToolUse, PostToolUse, UserPromptSubmit) to ~/.claude/settings.json, then speaks a short summary of what the agent did instead of reading the whole reply back. Voices come from local Kokoro (free, on your Mac) or your own ElevenLabs or Speechify key. Hooks return immediately, so Claude Code never waits on speech.
Want something smaller? Three single-purpose repos from the same team:
/say skill (MIT) that reads a short summary aloud with your computer's built-in voice. No API key, no account.To make any other command-line AI agent talk, put heard run in front of the command you normally use:
heard run aider
heard run your-agent --project ./my-project
Heard starts the process under a PTY, passes its output through, and speaks what it prints. Quality depends on what the process emits: full-screen terminal interfaces may not give it enough to work with, and questions or approvals may still need the agent's own interface. Every wrapped command shares one session. heard run is part of this open source engine (see Self-host), not the Mac app. Claude Code and Codex don't need it: heard install claude-code and heard install codex give richer events. More in the docs.
The engine in this repo is open source under Apache 2.0 and needs no account or API keys. The voice runs on your Mac; the list below says exactly what can leave it:
heard download-voice once (about 350 MB).claude command is installed, the engine can send agent activity to Anthropic through claude -p, with your Claude login, to write rolled-up project summaries. An ambiguous reply to Heard's catch-up prompt can go the same way.product_analytics: false in the config file that heard config path prints.git clone https://github.com/heardlabs/heard.git && cd heard
python3 -m venv .venv && source .venv/bin/activate && pip install -e .
heard download-voice # local voice, about 350 MB, one time
heard install claude-code # or: heard install codex
Full options in Self-host. Comparing local voice tools for coding agents? See open-source voice tools for coding agents.
🔊 Turn sound on for demo
https://github.com/user-attachments/assets/d823a946-fb6f-438b-904f-aa66d4268ed1
The app is the managed experience: download, sign in, and you're narrating. Heard runs the cloud voices and the narration brain for you, with no keys to manage. Every new account starts with a 14-day Power trial, no card required. Prefer to run it yourself with your own keys? See Self-host.
The app also adds cloud agents like Grok Bot, Meta Muse, Devin, Manus and ChatGPT, and on Power you talk back.
Heard gives coding agents a voice. Heard (heard.dev) is the voice context layer for agents.
Paste this into Claude Code, Codex, or any AI coding tool:
Install Heard so you can narrate your work to me out loud. Run:
curl -L https://heard.dev/download/cc -o /tmp/heard.zip && unzip -o /tmp/heard.zip -d /Applications && xattr -dr com.apple.quarantine /Applications/Heard.app && open /Applications/Heard.app, then hand it back to me - a quick setup window opens and I'll take it from there.
brew install --cask heardlabs/heard/heard
Download Heard.dmg, drag Heard.app into /Applications, and open it. Setup walks you through sign-in, permissions and your voice.
Heard listens to your agent, not your terminal, so narration works wherever the agent runs: Ghostty, Herdr, cmux, iTerm2, Terminal, Warp, WezTerm, tmux, VS Code, Cursor, Windsurf, Zed. Heard also knows where each session lives, so a jump back lands on the exact Ghostty terminal or tmux pane, or the right iTerm2 or Terminal window.
Agents that live in the cloud can't reach your Mac, so they report to Heard instead. Add one remote MCP server to the agent, https://api.heard.dev/v1/mcp/agent, and its progress, questions and results are spoken on your Mac like any local session, named after the agent: "Scout finished: …", "Muse needs you: …".
| In Settings | Agents |
|---|---|
| Authorize (sign in with your Heard email) | Grok Bot, ChatGPT, Claude |
| Token (created in Heard) | Meta Muse, Devin, GitHub Copilot coding agent, Manus |
Set these up in Settings → Connections → Cloud agents: each agent has numbered steps and the exact text to paste, each with a Copy button. More agents connect the same way through the CLI bundled with the app: Cursor, Instinct, OpenAI Dots, Claude Code or Codex running on another machine, or any other MCP-capable agent:
/Applications/Heard.app/Contents/MacOS/heard-board connect <agent>
Talk back (Power): hold right ⌘ and start with the agent's name, "Muse, summarize my inbox". Heard queues the message in that agent's Heard inbox, the agent picks it up with the heard_inbox tool between steps, and Heard tells you when it did. Or ask Parrot: *"Hey Parrot, tell Scout to