Embedded Video Captions
heygen-com/hyperframes
Adds captions to a single-subject talking-head video without editing the footage, from plain subtitles to cinematic text placed behind the speaker.
Produces a local rough cut of a talking-head or narrated screen recording, with transcript correction, an approval gate, pause compression and loudness normalization.
Install Local Talking-Head Rough Cut
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cut --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ra-local-talking-head-cut .claude/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .claude/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cutType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cut --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/ra-local-talking-head-cut .agents/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .agents/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cut --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/ra-local-talking-head-cut .cursor/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .cursor/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/Pluviobyte/rnskill.git --path skills/ra-local-talking-head-cut--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cut --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/ra-local-talking-head-cut .gemini/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .gemini/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cutInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/ra-local-talking-head-cut .github/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .github/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install Pluviobyte/rnskill ra-local-talking-head-cut --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Pluviobyte/rnskill.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/ra-local-talking-head-cut .opencode/skills/ra-local-talking-head-cut && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "ra-local-talking-head-cut" agent skill from https://github.com/Pluviobyte/rnskill/tree/main/skills/ra-local-talking-head-cut into .opencode/skills/ra-local-talking-head-cut/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ra-local-talking-head-cut", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
ra-local-talking-head-cut- GitHub stars
- 1.6k
- Token cost
- ~1.8k tokens
- SKILL.md length
- 636 words
- Files
- 11 (incl. scripts, references)
- Skills in repo
- 45
- Repo updated
- First seen
- Licence
- Custom licence
At a glance
Produces a local rough cut of a talking-head or narrated screen recording, with transcript correction, an approval gate, pause compression and loudness normalization.
- Works in 12 steps: Probe the source. Preserve width,… → Transcribe with the installed… → Run scripts/prepare_transcript.py with → …
- Cleaning up a talking-head recording by removing failed takes and long pauses
- SKILL.md covers Required workflow, Commands and Decision gates
- Runs Python scripts from its folder; calls python3
What it does
This skill is a reproducible local workflow for editing a talking-head or screen-recording video without a cloud editor, aimed at Chinese or mixed Chinese-English speech. It probes the source to keep width, height and frame rate, transcribes with the video-use transcribe helper using word timestamps, then runs scripts/prepare_transcript.py with a default glossary. That produces a review document, a pre-cut SRT, a corrected script, a list of uncertain terms and a subtitle approval file.
The agent must stop and wait for your approval of the corrected terms before building an edit list, and regenerating the files resets that approval. After approval, scripts/build_edl.py refuses to run until subtitles are approved, keeps pauses up to 550 ms and compresses longer ones to between 380 and 450 ms. scripts/render_cut.py then renders a hard-cut preview with 15 ms audio fades and light dialogue cleanup, and normalizes loudness to -16 LUFS. Other scripts generate final subtitles, burn captions, analyze visual cuts and run quality checks.
When your agent uses it
- Cleaning up a talking-head recording by removing failed takes and long pauses
- Correcting product names in a transcript before editing by meaning
- Making final-audio subtitles for a narrated screen recording
- Normalizing dialogue loudness while keeping the source resolution and frame rate
Example prompts
- “Make a rough cut of demo.mp4, correct the product terms and wait for my approval before cutting.”
- “Compress the long pauses in this narrated screen recording without making the speech sound rushed.”
- “Generate subtitles that match the final audio of the approved cut and burn them in.”
Requirements
- The video-use skill's transcribe helper installed
- Python to run the bundled scripts
Workflow steps
12 steps, taken from the first numbered list in SKILL.md.
- Probe the source. Preserve width, height, and frame rate unless the user
- Transcribe with the installed video-use/helpers/transcribe.py using
- Run scripts/prepare_transcript.py with
- Give the user the source video plus precut-review.srt and the readable
- Resolve uncertain terms by adding confirmed mappings to a job-local copy of
- Write optional decisions.json for semantic deletions. Delete repeated
- Run scripts/build_edl.py with the approval file. The script must refuse to
- Run scripts/render_cut.py without transitions to create a hard-cut preview.
- Run scripts/analyze_visual_cuts.py on the hard-cut preview. Inspect its
- Run scripts/qc.py against the chosen clean final MP4 and pass it the same
- Run scripts/generate_final_subtitles.py against the exact final MP4 and
- When the deliverable needs visible subtitles, run skill-captions with
What it can do on your machine
Read from SKILL.md and the folder at commit 83d1783. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 7 files in scripts/ (Python), which the agent can run.
Shell commands in SKILL.md call:
python3
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Local Talking-Head Rough Cut loads about 1.8k tokens when it runs, and up to ~2.9k if it reads all its reference files. Until then it costs about 120 tokens; SKILL.md has 636 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 636 words (~1,777 tokens).
“Deliver a reproducible local workflow, not an editing product. Keep upstream skills untouched and write every job under the source project's engineering directory.”
- name
- ra-local-talking-head-cut
Files
SKILL.md and 10 other files (scripts, references) in skills/ra-local-talking-head-cut of Pluviobyte/rnskill.
- SKILL.md
- agents/openai.yaml
- references/artifact-contract.md
- references/default-glossary.json
- scripts/analyze_visual_cuts.py
- scripts/build_edl.py
- scripts/burn_captions.py
- scripts/generate_final_subtitles.py
- scripts/prepare_transcript.py
- scripts/qc.py
- scripts/render_cut.py
Open the folder on GitHubat commit 83d1783
Compare with similar skills
Local Talking-Head Rough Cut next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Local Talking-Head Rough Cut this skillPluviobyte/rnskill | 1.6k | — | ~1.8k | Automated safety check: Pass | Custom licence | |
| Embedded Video Captionsheygen-com/hyperframes | 60k | 3 repos | ~8.6k | Automated safety check: Pass | Apache-2.0 | |
| Video Understandcalesthio/OpenMontage | 66k | — | ~841 | Automated safety check: Pass | AGPL-3.0 | |
| VideoCaptioner SubtitlesWEIFENG2333/VideoCaptioner | 16k | — | ~1.7k | Automated safety check: Pass | GPL-3.0 | |
| KrillinAI CLI Operatorkrillinai/OpenCreator | 13k | — | ~869 | Automated safety check: Pass | Apache-2.0 | |
| Conversational Video Editingbrowser-use/video-use | 29k | 2 repos | ~6.4k | Automated safety check: Warn | MIT |
Similar skills
Video Understand
calesthio/OpenMontage
Understand video content locally using ffmpeg frame extraction and Whisper transcription.
VideoCaptioner Subtitles
WEIFENG2333/VideoCaptioner
Adds subtitles to video with the videocaptioner CLI: transcribes speech, tidies and translates the text, and burns styled subtitles into the video or exports SRT and ASS files.
KrillinAI CLI Operator
krillinai/OpenCreator
Routes agents to the right KrillinAI command for subtitles, dubbing, video rendering, covers and speech, and explains how to read its JSON and manifest output.
Conversational Video Editing
browser-use/video-use
Edits video through conversation: transcribes speech, proposes cuts, color grades, adds overlay animations and burns in subtitles, confirming the plan before each edit.
Capcut Edit
renezander030/capcut-cli
Edit CapCut / JianYing video projects — read and write subtitles, timing, speed, volume, templates, animations (fade/ken-burns), and cut long-form to shorts.
More from Pluviobyte/rnskill
All 45 skills in this repoMotion Replica Builder
Pluviobyte/rnskill
Analyzes a reference video's layout, motion and timing over a chosen time range, then builds an original animated replica with new copy and assets plus a final video QC.
Niulai Style Image Transfer
Pluviobyte/rnskill
Restyles photos, film stills or new scene descriptions in the rough low-budget 3D look observed in the 2026 animated film Niulai, with an iterative scoring loop.
Skill Captions
Pluviobyte/rnskill
Renders, previews, validates and burns in production subtitles from QC-passed caption timing, with fixed dark or light caption styles scaled to the video size.
Chinese Social Media Cover Maker
Pluviobyte/rnskill
Creates Chinese social-media video covers in 3:4 and 4:3 from registered styles and shared presenter assets, or swaps the person in a reference cover for a digital-human frame.
HeyGen Digital Avatar Compositor
Pluviobyte/rnskill
Generates a HeyGen digital-human video layer using a fixed avatar profile, local narration and an approved circular avatar placement.
Audio-Timed Subtitle Generator
Pluviobyte/rnskill
Generates SRT, VTT and caption JSON for final narration audio or a merged video using Volcengine Doubao ASR word timestamps, with a quality gate before delivery.
Related
Categories
Questions about Local Talking-Head Rough Cut
What does Local Talking-Head Rough Cut do?
Produces a local rough cut of a talking-head or narrated screen recording, with transcript correction, an approval gate, pause compression and loudness normalization. This skill is a reproducible local workflow for editing a talking-head or screen-recording video without a cloud editor, aimed at Chinese or mixed Chinese-English speech.py with a default glossary.
When should I use Local Talking-Head Rough Cut?
Local Talking-Head Rough Cut fits situations like: cleaning up a talking-head recording by removing failed takes and long pauses; correcting product names in a transcript before editing by meaning; making final-audio subtitles for a narrated screen recording; normalizing dialogue loudness while keeping the source resolution and frame rate.
How do I install Local Talking-Head Rough Cut in Claude Code?
Run `npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a claude-code`. Or copy the skill folder (skills/ra-local-talking-head-cut in Pluviobyte/rnskill) into .claude/skills/ra-local-talking-head-cut in your project. Claude Code loads it when a task matches its description.
How do I install Local Talking-Head Rough Cut in Codex?
Run `npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a codex`. Or copy the skill folder (skills/ra-local-talking-head-cut in Pluviobyte/rnskill) into .agents/skills/ra-local-talking-head-cut in your project. Codex loads it when a task matches its description.
Can I use Local Talking-Head Rough Cut in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Pluviobyte/rnskill --skill ra-local-talking-head-cut -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ra-local-talking-head-cut, .gemini/skills/ra-local-talking-head-cut, .github/skills/ra-local-talking-head-cut and .opencode/skills/ra-local-talking-head-cut in your project.
What does Local Talking-Head Rough Cut need to run?
Going by SKILL.md and its folder, Local Talking-Head Rough Cut needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: The video-use skill's transcribe helper installed; Python to run the bundled scripts.
Does Local Talking-Head Rough Cut access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Local Talking-Head Rough Cut safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Local Talking-Head Rough Cut use?
Local Talking-Head Rough Cut has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
How many tokens does Local Talking-Head Rough Cut use?
About 1.8k tokens (SKILL.md is roughly 7.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.2k tokens, read only when the agent opens those files.
What are the alternatives to Local Talking-Head Rough Cut?
Skills that share tags, products or a category with Local Talking-Head Rough Cut: Embedded Video Captions (heygen-com/hyperframes, 60k stars), Video Understand (calesthio/OpenMontage, 66k stars), VideoCaptioner Subtitles (WEIFENG2333/VideoCaptioner, 16k stars) and KrillinAI CLI Operator (krillinai/OpenCreator, 13k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Local Talking-Head Rough Cut?
Pluviobyte (a GitHub user) maintains it in Pluviobyte/rnskill, which has 1,639 GitHub stars. The repository holds 45 skills in this directory. The repository was last updated on September 21, 2026.
Source: Pluviobyte/rnskill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.