video-editor-agent

In the author's own words, from each skill's SKILL.md.

• ai-audio-sound-design

  Rebuild the audio of an AI-generated video (Arcads / AI actors / UGC-gen / any clip whose voices sound like sterile studio AI) so every scene sounds like it was really filmed in the location shown — location-matched ambience beds, room-matched convolution reverb, outdoor distance treatment, censor bleeps over swears, removal of the AI high-frequency watermark whine, and a social loudness master,…

  .claude/skills/ai-audio-sound-design/SKILL.md on GitHub (opens in a new tab)

• arcads-broll

  GENERATE B-roll and motion-graphic elements for an edit with AI video/image models via the Arcads external API (Seedance, Sora 2, Veo 3.1, Kling, Grok Video, Nano Banana) — product b-roll clips, scene beats, UGC-style inserts, shot-card stills, and screen-blend motion-graphic overlays. Use when an edit needs footage that cannot be captured: "generate b-roll of the product", "make an AI clip for t…

  .claude/skills/arcads-broll/SKILL.md on GitHub (opens in a new tab)

• arcads-video-edit

  Turn a creator's multi-take product-demo recordings (each take = screen recording + camera video + mic WAV, usually the creator walking through Arcads) into a finished 9:16 ad: an EDL-driven base cut rotating three layouts (the speaker full screen / split with the screen on top / full-screen product with the speaker in a circle), every pause cut so they talk over their own pauses, retakes dropped…

  .claude/skills/arcads-video-edit/SKILL.md on GitHub (opens in a new tab)

• branded-ad-edit

  Turn a RAW talking-head video into a FINISHED, branded, motion-graphics ad for any brand or industry — a designed motion graphic on every spoken line, the brand's real colors/logo/screenshots, word-synced karaoke captions, ElevenLabs SFX + a low music bed, website B-roll in browser frames, dramatic full-bleed cuts, and optional hook A/B variants. Renders 9:16 MP4s via HyperFrames. Use this whenev…

  .claude/skills/branded-ad-edit/SKILL.md on GitHub (opens in a new tab)

• broll-capture

  Capture website B-roll and screenshots for a video edit: full-page screenshots of real pages (for shot-cards and "proof" beats), smooth scripted-scroll screen recordings, and high-fidelity Screen Studio recordings of the user's real browser for sites that block automation. Use whenever an edit needs real screenshots, a screen recording of a website/app, "capture the site", or B-roll of a product.…

  .claude/skills/broll-capture/SKILL.md on GitHub (opens in a new tab)

• capcut-export

  Export a finished layered edit (footage + graphic cards + alpha fx overlay + captions + SFX/music/VO) into a CapCut desktop draft where EVERY element stays editable — footage segments split at zoom boundaries, per-card graphic clips, a true-alpha fx overlay, native text captions, per-hit SFX clips, and separate music/VO tracks. Use when someone says "export to CapCut", "make this editable in CapC…

  .claude/skills/capcut-export/SKILL.md on GitHub (opens in a new tab)

• edl-tighten

  Surgically remove awkward pauses, dead air, and long silences from a FINISHED voiceover-driven cut while keeping video, VO audio, word-synced captions, SFX hits, and the composition timeline in sync. Use when asked to "tighten the cut", "remove the pause at 0:41", "cut the dead air", "the gap between sentences is too long", "shave the silences", "it drags here", or when a revision round adds new…

  .claude/skills/edl-tighten/SKILL.md on GitHub (opens in a new tab)

• hook-splitter

  Split ONE long recording that contains many separate hooks/takes into a set of standalone videos — one per hook — and tighten each so there is no dead air, no stumble, and nothing said twice. Use whenever the user hands over a single long file and wants the pieces cut out: "split these hooks out", "there are 20+ hooks in here, make them each their own video", "cut these into individual clips", "o…

  .claude/skills/hook-splitter/SKILL.md on GitHub (opens in a new tab)

• hook-variations

  Combine ONE finished body cut with MANY hook clips to produce one standalone video per hook — the hook A/B batch you upload as separate ad creatives. Use whenever someone wants to "add all these hooks to this body", "combine this video with every hook in that folder", "make 21 versions with different openers", "swap the hook", "build hook variants", "test these hooks against the same body", or ha…

  .claude/skills/hook-variations/SKILL.md on GitHub (opens in a new tab)

• naming-convention

  Name a batch of video deliverables so each filename says what the file actually is — subject, variant axis, and treatment — instead of `render-final-v3.mp4`. Use whenever more than two files are being handed over, and whenever someone says "rename these", "give these descriptive names", "I can't tell these apart", "name them so I know which is which", "put the product in the filename", "which one…

  .claude/skills/naming-convention/SKILL.md on GitHub (opens in a new tab)

• openart-broll

  GENERATE B-roll, talking-head clips, and screen-blend motion-graphic overlays for an edit through the OpenArt MCP (Seedance, Kling, Wan, Gemini Omni, Grok Imagine, Nano Banana and more) — identity-referenced talking heads from a reference video, product/scene beats, and flat-black overlay clips generated per spoken line. Use when an edit needs generated footage or overlays and the OpenArt MCP is…

  .claude/skills/openart-broll/SKILL.md on GitHub (opens in a new tab)

• recap-video

  Edit an event, conference, summit, trip, or behind-the-scenes recap from spoken takes and supplied B-roll. Choose the strongest take for each line, remove pauses and restarts, match footage to the narration, mix full-screen and split layouts, add captions and selective graphics, then render, verify and revise. Use for "recap video", "event recap", or "edit this summit like this reference"; not fo…

  .claude/skills/recap-video/SKILL.md on GitHub (opens in a new tab)

• reel-recut

  Recut a raw vertical talking-head reel into a finished short-form edit: a persistent title banner, word-synced karaoke captions, callout boxes for asides and comment-keyword CTAs, and tight silence-cut "talking over myself" pacing — all described in ONE JSON spec and rendered deterministically. Use when the user supplies a talking-head video and wants it edited, styled, or "made to match my other…

  .claude/skills/reel-recut/SKILL.md on GitHub (opens in a new tab)

• reel-style-clone

  Reverse-engineer a reference reel's editing style frame-by-frame into a replicable STYLE GUIDE plus per-line BUILD DIRECTIVES, then apply that style to the user's own footage. Use whenever someone hands over (or points at) an existing short-form video and says "edit mine like this", "clone this editing style", "match this reel", "why does this edit feel so good — copy it", "break down how this vi…

  .claude/skills/reel-style-clone/SKILL.md on GitHub (opens in a new tab)

• sound-design

  Generate and mix the full audio pass for a video edit with ElevenLabs — an SFX kit (whooshes, pops, impacts, count-ups, sub-drops, risers) generated, audited for the ~1/3 near-silent duds, and peak-normalized to −3dB; a music bed prompted by style descriptors (optionally style-matched to a reference track via spectrogram + BPM estimation); and the proven mix levels + ffmpeg verification recipe to…

  .claude/skills/sound-design/SKILL.md on GitHub (opens in a new tab)

• talking-head-image-overlays

  Edit a talking-head video in the "uncut take + stacked image overlays" style: ONE continuous take that is never cut, with the entire sense of pace carried by hard-cut overlays landing on content nouns — a short serif title plate, found-footage rectangles above the eyeline, opaque full-frame skeleton-UI cards, raw prompt text on black, word-by-word karaoke captions, and a CTA banner that arrives e…

  .claude/skills/talking-head-image-overlays/SKILL.md on GitHub (opens in a new tab)

• video-edit-pipeline

  MASTER orchestration skill for producing a finished short-form video from raw talking-head footage. Use for ANY request to edit, produce, recut, or "make an ad/reel out of" talking-head footage — with or without a style reference — including "edit this video", "edit this like the reference reel", "turn this clip into a finished reel", "do the full edit", "make this look like their winning ads", o…

  .claude/skills/video-edit-pipeline/SKILL.md on GitHub (opens in a new tab)

• video-qa

  Four-layer QA for a rendered video edit — run it on any .mp4/.mov you or a pipeline just produced. Use when asked to QA / check / verify / audit a render or export, hunt clipped words, duplicate dialogue at cuts, dead air, splice clicks, black / frozen / flash frames, loudness (LUFS) or clipping problems, confirm a fix after a re-render, or build an inspection packet (contact sheet + waveform + t…

  .claude/skills/video-qa/SKILL.md on GitHub (opens in a new tab)

• video-review-canvas

  Publish a finished (or draft) video cut to a frame.io-style review canvas on here.now — a player with a frame-accurate scrubber, persistent timeline notes, version comparison, replies, resolution and evidence, and a clickable beat map — then read those notes back before the next cut. Use this to DELIVER any video edit for review: reel recuts, brand-deal edits, motion-graphics ads, carousel videos…

  .claude/skills/video-review-canvas/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• ECCby affaan-m The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Cod… 280k stars · 30 skills
• reaby morluto Reverse engineer anything with agents, from app behavior down to native binaries. 270k stars · 1 skill
• andrej-karpathy-skillsby multica-ai A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. 220k stars · 1 skill
• ponytailby DietrichGebert Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. 164k stars · 6 skills
• ui-ux-pro-max-skillby nextlevelbuilder An AI skill that provides design intelligence for building professional UI/UX across multiple platforms. 136k stars · 7 skills
• graphifyby Graphify-Labs Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Curs… 127k stars · 1 skill