heard

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.

Runssource: GitHubhomepagePythonApache-2.0commit c82958d08a77

Python, Apache-2.0 licensed. The project labels itself: agent notifications, ai agents, claude code, claude code hooks, codex, codex cli, coding agents and developer tools.

heard runs. An Argusic agent installed it in 3 minutes and hit 3 errors on a clean machine with no GPU, and the whole session was recorded.

Heard Python package v1.1.48 installed and importable; CLI functional; 871/872 relevant tests pass on Linux (the 1 failure is macOS-specific afplay binary missing); ruff lint clean

What the agent ended up with on a clean machine, in its own words. How this is measured

Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 86.7 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs
Argusic Score
86.7 / 100
cost of the verifying run
$0.05 (measured)
recorded runs
1
last tested
stars
438
forks
31
open issues
8
watchers
9
size
3 MB
created
last push

Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.

What was measured

From the run that decided this verdict. Every line is a recorded value, not an estimate.

Time lost to each failure while testing heard, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • pyobjc-framework-* (macOS-only deps) failed to build on Linux - sw_vers not found1 minute
  • test_muted.py::test_speak_drops_when_mic_active fails - afplay not available on Linux
  • test_in_app_update.py::test_unzip_app_extracts_bundle fails - unzip not available on Linux
Install time
3 minutes
Cold machine to finish
3 minutes
Errors hit and fixed
3 hit, 1 fixed with no human help
How the result was proved
heard --help exited 0 showing 5 commands; heard run echo 'hello world from heard' printed output; python -c 'import heard; print(heard.__version__)' returned 1.1.48; pytest --ignore=tests/test_onboarded_resolution.py --ignore=tests/test_in_app_update.py -k 'not test_hotkey and not test_accessibility and not test_notify and not test_prompt_window and not test_ui and not test_onboarding' --tb=line exited 0 with 871 passed, 1 failed (afplay missing on Linux); ruff check heard/ tests/ passed
Model tokens used
134,258
Exact commit tested
c82958d08a77
Container image
sha256:cdd920bce783

What was not measured

The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.

Peak GPU memory
this ran on a CPU machine, so there was no GPU to sample
Behaviour over a long run
each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

tokens used
134,258
Here's the full report.
**Summary:** This is a macOS-specific app (py2app menu bar, `afplay`, Accessibility APIs) running o…
**Error details:**
| Error | Fix | Time |
|---|---|---|
| PEP 668 external env | Created venv | <1 min |
| pyobjc-framework-* can't build (sw_vers not found) | Used `--no-deps` install | <1 min |
| hatchling missing `editables` / `setuptools` | Installed missing deps | <1 min |
| `afplay` not found in test | Not fixable (macOS binary) | 0 min |
| `unzip` not found in test_in_app_update.py | Not fixable (Linux) | 0 min |

Replay the whole session, every command from a clean machine to this point.

How it was tested

One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.

Strengths and limits

Measured facts, not opinions. How this is written.

What went well

  • Reached a running state on a clean machine, with the session recorded.
  • Installed in 3 minutes.
  • Ran without a GPU, so it does not need one to start.
  • Licensed Apache-2.0, as reported by its host.

What did not, or is not known

  • Hit 3 errors during setup, 2 of which were not solved.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

What it is a reasonable choice for

  • Trying it on a laptop or a small server: it reached a running state without a GPU.
  • A quick evaluation: it was running 3 minutes after the clone.

Also tested, in the same area

Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs86.670.05

Topics (from GitHub)

agent-notificationsai-agentsclaude-codeclaude-code-hookscodexcodex-clicoding-agentsdeveloper-toolshands-freekokorolocal-firstmacosofflineopen-sourceread-aloudtext-to-speechttsvoicevoice-assistantvoice-output

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/heard.svg)](https://argusic.com/subject/heard)

Questions

Does heard run?
Yes. heard runs. Argusic installed and launched it on a clean machine in 3 minutes, hitting 3 errors on the way, and recorded the session.
How did Argusic test heard?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit c82958d08a77. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test heard?
The run that produced this verdict cost $0.05: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does heard take to install?
3 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
Does heard need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing heard?
3 things broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
Where is the evidence for heard?
The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion