ghost-editor
Raw talking-head take in, finished reel out. Open-source AI video editor for Claude Code.
Runs with mockssource: GitHubPythonMITcommit ce172726c362
Python, MIT licensed. The project labels itself: agent skills, ai video editor, capcut alternative, captions, claude code, claude skills, codex and content creation.
ghost-editor runs, with stand-ins for the services it depends on. An Argusic agent installed it in 55 minutes and hit 5 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Install green: dependencies in ~/.venvs/media and Node 22, sfx/music/emoji-optional media restored, doctor reports 17 ok 0 missing against a local Gemini mock, hyperframes CLI answers 0.8.159, and the repo's test suite passes except for one pre-existing test-fixture assertion in test_cutout_fx.py that was deliberately left untouched.
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 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page. 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.
- tests/test_cutout_fx.py assertion: last --release frame differs from source by 20.68 gray levels (test requires <1.5)14 minutes
- python3 scripts/library_restore.py --sfx-only --music-only silently skipped both sections (flags are mutually exclusive in the script)12 minutes
- system python3 is externally managed (PEP 668), pip install refused10 minutes
- GEMINI_API_KEY required by doctor and QA scripts (real service)8 minutes
- ffmpeg/hyperframes need Node >= 22, system has Node 18.19.1; scripts/hf.sh --version failed with 'require Node.js >= 22'6 minutes
- Install time
- 55 minutes
- Cold machine to finish
- 30 minutes
- Errors hit and fixed
- 5 hit, 5 fixed with no human help
- How the result was proved
- PATH conversion: ~/node22/bin, MEDIA_PY=~/.venvs/media/bin/python, WHISPER=~/.venvs/media/bin/whisper, GOOGLE_GEMINI_BASE_URL=http://127.0.0.1:8765, GEMINI_API_KEY=mock-key. bash scripts/hf.sh --version -> 0.8.159; bash scripts/doctor.sh -> 17 ok, 0 missing (GEMINI_API_KEY works (one test call)); tests/test_assemble_sync.py -> PASS with 744 frames and 20/20 sync; tests/test_decode_levels.py -> ALL OK; ffmpeg -v error -i rel-1.mp4 -f null - -> decode ok; tests/test_cutout_fx.py -> all checks ok except last-frame mean update 20.678646 (limit 1.5)
- Model tokens used
- 1,206,122
- Exact commit tested
- ce172726c362
- 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
- Behaviour against real services
- this run was verified against stand-in services only
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- Produced media decoded cleanly: `ffmpeg -v error -i <test outputs> -f null -` → decode ok; h264 5… **Errors** 1. Node 18 too old for HyperFrames (>=22 needed) → pre-built Node 22 binary in `~/node22` → fixed. 2. system pip "externally managed" → user venv `~/.venvs/media` → fixed. 3. `library_restore.py --sfx-only --music-only` silently skipped both (mutually exclusive flags) → … 4. `scripts/hf.sh --version` failed without Node 22 on PATH → no repo change; a new user must use n… 5. `tests/test_cutout_fx.py` one assertion (20.68 vs 1.5) → root-caused as fixture/assertion issue;… exec /bin/sh -lc true in /work/repo succeeded in 0ms: tokens used 1,206,122
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
- Recovered from all 5 errors without a human stepping in, which says the failures are documented well enough to solve.
- Licensed MIT, as reported by its host.
What did not, or is not known
- Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
- Took 55 minutes to install, slower than the median of the 108 comparable projects Argusic has measured.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 109 agent-skills projects Argusic has installed and timed, ghost-editor was the 109th fastest to reach a running state, and 80 of 109 reached one at all.
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 by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.29 |
Topics (from GitHub)
agent-skillsai-video-editorcapcut-alternativecaptionsclaude-codeclaude-skillscodexcontent-creationhyperframesinstagram-reelsmotion-graphicsopus-clip-alternativereelsshort-form-videosubmagic-alternativesubtitlestalking-headtiktokvideo-editingyoutube-shorts
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/ghost-editor)Questions
- Does ghost-editor run?
- ghost-editor runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 55 minutes, hitting 5 errors on the way, and recorded the session.
- How did Argusic test ghost-editor?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit ce172726c362. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test ghost-editor?
- The run that produced this verdict cost $0.29: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does ghost-editor take to install?
- 55 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 ghost-editor need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing ghost-editor?
- 5 things broke in the recorded run, and 5 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does ghost-editor compare with the alternatives?
- Of the 109 agent-skills projects Argusic has installed and timed, ghost-editor was the 109th fastest to reach a running state, and 80 of 109 reached one at all.
- Where is the evidence for ghost-editor?
- 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.