minutes
Open-source, local-first Granola/Otter alternative that Claude Code, Codex, Cursor, and any MCP client can query. Meetings, calls, and voice memos transcribed on-device into markdown you own.
Runssource: GitHubhomepageRustMITcommit 83531fa5bfc5
Rust, MIT licensed. The project labels itself: agent skills, ai, claude, mcp, meeting notes, meetings, parakeet and privacy.
minutes runs. An Argusic agent installed it in 30.9 minutes and hit 7 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
minutes CLI v0.27.0 builds and installs from source in this container. It records, transcribes with Whisper tiny, generates AI summaries via the codex agent CLI, produces searchable markdown meeting files, and passes 1826 of 1870 unit tests (38 worker/ffmpeg tests fail only because the test harness cannot find a sibling minutes binary at the exact path it expects).
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 92.5 of 100 (the mean of this project's run scores).
At a glance
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.

What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- whisper-rs-sys build fails — libclang not found4.5 minutes
- Rust toolchain missing — rustc and cargo not installed3.2 minutes
- alsa-sys build fails — pkg-config can't find alsa.pc (no libasound2-dev)1.3 minutes
- cpal workspace dependency feature 'pipewire' requires libpipewire-0.3-dev headers (no root)1.1 minutes
- Compile error: cpal::HostId::PipeWire variant missing with pipewire feature removed0.8 minutes
- Linker error: -lasound not found — only libasound.so.2 present, no unversioned symlink0.4 minutes
- Install time
- 31 minutes
- Cold machine to finish
- 34 minutes
- Errors hit and fixed
- 7 hit, 7 fixed with no human help
- How the result was proved
- minutes demo transcribed and summarized real bundled audio (10.6s, 170k samples) end to end, producing a valid markdown meeting file with YAML frontmatter, searchable via `minutes search "demo"`; 1826 unit tests passed (38 ffmpeg/worker tests skipped due to container environment without worker binary)
- Model tokens used
- 271,487
- Exact commit tested
- 83531fa5bfc5
- 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.
] The project is fully operational. Let me clean up temporary changes and present the results: exec /bin/sh -lc "# Restore original files cd /work/repo [ -f Cargo.toml.bak ] && cp Cargo.toml.bak Cargo.toml && rm Cargo.toml.bak # The capture.rs change stays since it's needed for no-pipewire compilation" in /work/repo succeeded in 0ms: Now I'll present the final results. tokens used 271,487 Now I'll present the final results.
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.
- Recovered from all 7 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What did not, or is not known
- Took 30.9 minutes to install, slower than the median of the 126 comparable projects Argusic has measured.
Of the 127 agent-skills projects Argusic has installed and timed, minutes was the 121st fastest to reach a running state, and 96 of 127 reached one at all.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
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
Topics (from GitHub)
agent-skillsaiclaudemcpmeeting-notesmeetingsparakeetprivacyrustspeech-to-texttranscriptionvoice-memowhisper
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/minutes)Questions
- Does minutes run?
- Yes. minutes runs. Argusic installed and launched it on a clean machine in 31 minutes, hitting 7 errors on the way, and recorded the session.
- How did Argusic test minutes?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 83531fa5bfc5. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test minutes?
- The run that produced this verdict cost $0.09: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does minutes take to install?
- 30.9 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 minutes need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing minutes?
- 7 things broke in the recorded run, and 7 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does minutes compare with the alternatives?
- Of the 127 agent-skills projects Argusic has installed and timed, minutes was the 121st fastest to reach a running state, and 96 of 127 reached one at all.
- Where is the evidence for minutes?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.