# AI use cases for Ffmpeg Skill (kajisho5/ffmpeg-skill) Repo: [kajisho5/ffmpeg-skill](https://github.com/kajisho5/ffmpeg-skill) | Use cases: 4 | Page: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/ ffmpeg-skill is a local video editing toolkit for coding agents, with 42 FFmpeg-based tools, a machine-readable contract and an MCP server. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Cut an interview into a 60-second Reels clip: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/cut-interview-to-60s-reels-clip/ - Turn a talking-head video into a captioned TikTok: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/render-talk-to-tiktok-with-captions/ - Export one edit to nine social platforms at once: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/export-one-edit-to-every-platform/ - Strip dead air and set loudness to -14 LUFS: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/remove-silences-and-normalize-loudness/ ## How to cut an interview into a 60-second Reels clip with Ffmpeg Skill Repo: [kajisho5/ffmpeg-skill](https://github.com/kajisho5/ffmpeg-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README An agent keeps chosen segments of a long interview, fits them to 60 seconds in 9:16 and exports a Reels file. **Who it is for:** A creator or editor with a long interview recording who needs a short vertical clip for Instagram Reels and has no time to edit by hand. **The problem:** Manual trimming, reframing and export for a vertical social clip takes time, and the result is easy to get wrong on duration or aspect ratio. **The outcome:** A Reels-ready MP4 that is about 60 seconds, 1080x1920, with a probe and a contact sheet the agent has checked. ### Steps 1. Install the skill with `npx ffmpeg-skill` and check the machine with `npx ffmpeg-skill doctor`. 2. Set `S=~/.claude/skills/ffmpeg-skill/scripts`, then measure the source with `python3 $S/probe.py input.mp4 --compact`. 3. Keep the chosen parts with `python3 $S/cut.py input.mp4 --segments 0:45-3:10,5:00-6:30`. (inferred, check the README) 4. Preview the fit with `python3 $S/fit.py input.mp4 --duration 60 --aspect 9:16 --dry-run`, then run it without `--dry-run` using `--fit crop`. 5. Export with `python3 $S/export.py input.mp4 --preset reels --json` and check the result with `check.py --platform reels`. ### Prompt to paste into your agent ```text Using the ffmpeg-skill repo, take and keep 0:45-3:10 and 5:00-6:30, make it exactly 60 seconds at 9:16 for Reels, and export it. Expected output: a Reels MP4 of about 60 seconds at 1080x1920. Verify the duration and resolution with probe.py before you report it done. ``` ### Caveat The segment times and the 60-second target are your choices, and the skill does not decide which parts of the interview are worth keeping. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/cut-interview-to-60s-reels-clip/ ## How to turn a talking-head video into a captioned TikTok with Ffmpeg Skill Repo: [kajisho5/ffmpeg-skill](https://github.com/kajisho5/ffmpeg-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README One render command crops to 9:16, burns in your cue text and normalizes loudness for TikTok. **Who it is for:** A creator with a talking-head recording and a text file of caption cues, who wants a finished TikTok without learning FFmpeg. **The problem:** Captioning, reframing, loudness and platform export are separate steps, and the caption text has to sit clear of the app's own UI. **The outcome:** A TikTok-ready file with popped captions above the description bar, around -14 LUFS, plus a check report written by the skill. ### Steps 1. Install the skill with `npx ffmpeg-skill` and confirm the tools are usable with `npx ffmpeg-skill doctor`. 2. Write your caption lines into a cues file, one cue per line, with the text as you want it burned in. 3. Run `python3 $S/render.py talk.mp4 --template tiktok --cues cues.txt`, after setting `S` to the skill's scripts folder. 4. Find the output next to the input, or use `-o` to choose another folder, and confirm the check.py result for TikTok. ### Prompt to paste into your agent ```text Using the ffmpeg-skill repo, render with the tiktok template and the cues file . Expected output: a 9:16 TikTok MP4 with captions and a platform check written next to it. Verify the check result and the output resolution before reporting success. ``` ### Caveat The cue text is burned in exactly as written, so the wording and line breaks are your responsibility. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/render-talk-to-tiktok-with-captions/ ## How to export one edit to nine social platforms at once with Ffmpeg Skill Repo: [kajisho5/ffmpeg-skill](https://github.com/kajisho5/ffmpeg-skill) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README A single render command makes a file for each platform template and a summary table. **Who it is for:** A marketer or creator who publishes the same video on TikTok, Reels, Shorts, LinkedIn, X and other platforms and wants all versions from one edit. **The problem:** Each platform needs its own aspect ratio, limits and export settings, and making each version by hand repeats the same work. **The outcome:** One output file per platform template, plus a pack table that lists what each destination produced. ### Steps 1. Install the skill with `npx ffmpeg-skill` and run `npx ffmpeg-skill doctor`. 2. List the templates and their frames, limits and safe zones with `python3 $S/render.py --list-templates`. (inferred, check the README) 3. Preview every planned command with `python3 $S/render.py talk.mp4 --template all --dry-run`. (inferred, check the README) 4. Render all destinations with `python3 $S/render.py talk.mp4 --template all`, then read the ` _pack.md` table it writes. (inferred, check the README) ### Prompt to paste into your agent ```text Using the ffmpeg-skill repo, render with --template all so it covers every platform template. Expected output: one file per platform plus a _pack.md summary table. Verify the summary lists every destination and that each output probes correctly. ``` ### Caveat Platform limits and safe zones come from the shipped templates, so check them against the current app rules before you publish. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/export-one-edit-to-every-platform/ ## How to strip dead air and set loudness to -14 LUFS with Ffmpeg Skill Repo: [kajisho5/ffmpeg-skill](https://github.com/kajisho5/ffmpeg-skill) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README An agent removes silent gaps from a recording and sets its loudness to the -14 LUFS target. **Who it is for:** A podcaster or presenter whose raw recording has long pauses and uneven volume and who wants a cleaned file before publishing. **The problem:** Finding and cutting silent gaps by hand is slow, and loudness levels often differ between recordings. **The outcome:** An edited file with silent gaps removed and integrated loudness near -14 LUFS with a true peak of -1 dBTP. ### Steps 1. Install the skill with `npx ffmpeg-skill` and check the machine with `npx ffmpeg-skill doctor`. 2. Remove silences above a chosen threshold with `python3 $S/silence.py recording.mp4 --threshold -35`. (inferred, check the README) 3. Normalize loudness with `python3 $S/loudness.py cleaned.mp4 -I -14 --tp -1`. (inferred, check the README) 4. Verify the output with `python3 $S/probe.py cleaned.mp4 --compact` before publishing. (inferred, check the README) ### Prompt to paste into your agent ```text Using the ffmpeg-skill repo, remove silences from with a -35 threshold, then normalize loudness to -14 LUFS with a -1 true peak. Expected output: a cleaned MP4 with the target loudness. Verify the result with probe.py and report the duration change. ``` ### Caveat The -35 dB threshold is only a starting point, so listen to the result and tune it for your recording. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kajisho5-ffmpeg-skill/remove-silences-and-normalize-loudness/