Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/belagum/promptcut/setupgit clone --depth 1 https://github.com/Belagum/promptcutWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00019 | $0.00432 |
| Opus 5 | $0.00010 | $0.00216 |
| Sonnet 5 | $0.00004 | $0.00086 |
| Haiku 4.5 | $0.00002 | $0.00043 |
Grade A, and why
setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Using the video-toolbox skill: run setup first - it pip-installs the missing
optional packages (pillow, yt-dlp, edge-tts, pycapcut) and ffmpeg via
winget/brew on its own. Then run doctor and walk the person through only what
is still red. Extra context: $ARGUMENTS
What setup cannot do by itself:
- OpenRouter key: openrouter.ai/settings/keys, then
config --set openrouter_api_key=sk-or-...(works immediately), orsetx OPENROUTER_API_KEY sk-or-...+ a new terminal. - Free voice instead of paid TTS:
config --set tts_provider=edgeand pick a voice, e.g.ru-RU-DmitryNeural. - Stock images and music: free keys from pexels.com/api, pixabay.com/api/docs,
freesound.org/apiv2/apply, devportal.jamendo.com, then
keys --set pexels=... freesound=.... Openverse, Wikimedia and archive.org need no key. - CapCut export:
capcut-drafts; if the folder is not found, ask for CapCut → Settings → Draft location. - Premiere / Resolve / VEGAS export needs nothing extra;
doctorlists the installed editors undernle. Only VEGAS--runmay needconfig --set vegas_exe=...when the exe is not auto-detected. - If ffmpeg was just installed by winget, a new terminal may be needed for PATH.
- If YouTube demands a login in
media-dl:config --set ytdlp_cookies=firefox(or chrome/edge, or a cookies.txt path).
Finish with a one-line summary of what works now and what is still missing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 31 lines · 19 tokens per session scan A f19fd6bad54d
setup is a command published in the GitHub repository Belagum/promptcut (1 stars, last pushed 6d ago), licensed MIT. It adds 19 tokens to every session and 432 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
generate-all
Generate creative assets for all posts in the current month's calendar.
edit-image
Edit a generated image — adjust background, lighting, colors, or composition.
index-assets
Index or re-index a brand's visual asset library.
render-carousels
Render multi-slide carousels from HTML templates via Playwright.
edit-post
Edit an existing post's copy, visual direction, or metadata.
generate-post
Generate creative for a single post by ID.