beatra-ai/beatra-skills / skills/ai-music-video-clip-maker
ai-music-video-clip-maker
VerifiedCreate a short visual clip guided by a song's mood, rhythm, and visual concept. This AI music video clip maker and song-to-video generator turns a music excerpt and visual direction into a cinematic music promo clip, animates approved cover art or a portrait in time with the music, interpolates motion between opening and ending art, or uses audio as a loose mood reference for a new visual concept. Use it for new-song teasers, album promo clips, cover art animation, mood visuals, virtual performer scenes, and social music teasers, with an audio-visual map built from the song's hook, energy, palette, and landing image.
AI Music Video Clip Maker
Create a short visual clip guided by a song's mood, rhythm, and visual concept. Use this Skill for AI music video clips, song-to-video teasers, new-song promos, album promo clips, cover art animation, mood visuals, virtual performer scenes, and social music teasers.
Scope and adjacent routes
The song is the fixed spine: the clip exists to amplify its structure, artist identity, lyrics, and release goal. If rights are not cleared, stay at planning or an internal sample and do not treat the result as a publishable official music video. Route a portrait-plus-singing-audio request to a photo-singing workflow; a story-to-lyrics-to-song request to a personalized-song workflow; a reference-song reinterpretation to a song-cover workflow; and a broad non-music video request to a general video studio. Keep this route focused on finished music becoming a short visual clip. Do not generate a song from a theme in this package.
Inputs and defaults
The hard inputs are:
- one finished song or accessible excerpt whose real duration can be cut to a supported video duration;
- one main route: performance, narrative, lyric-led, or visualization.
Ask only for a missing hard input. Reuse the known genre, energy, opening image, and aspect ratio. A cover or opening still is not a hard input. When the visual direction exists and no usable cover or opening still is present, generate one cover or keyframe with beatra.images.generate as its own paid stage. For a local audio or image file the host Agent can access, use the bundled upload helper only after inspection:
python3 scripts/mcp_client.py upload ./song-excerpt.mp3 --mime-type audio/mpeg
python3 scripts/mcp_client.py upload ./cover-art.png --mime-type image/png
Upload is transport, not creative review. Retain the returned artifact reference and never pass a local path to a remote tool.
Default to cutting the chorus or strongest 20–30 seconds rather than forcing the whole song into one clip. When this account has the music tools (the bundled client's tools command lists beatra.music.edit and beatra.audio.process; see music editing and audio), make that cut with beatra.audio.process operation: "trim" (start_ms, and end_ms or target_duration_ms; a paid call at the live audio_process price, on its own card with its own client_request_id) instead of asking the user for a new excerpt: take the section's start and end from the song's lyric timing when one was made, otherwise ask the user where it starts and ends in minutes and seconds as their player shows it ("chorus 1:05 to 1:35"), convert that to start_ms/end_ms, and name the part kept; without the music tools, work from the excerpt the user supplies. Keep model: "auto" and a source-appropriate aspect ratio. Build a compact audio-visual map before any paid call: hook or energy, dominant mood, pulse, visual motif, subject, movement, palette, camera, opening image, and landing image. Write the beat, phrase, or climax cut into the paid card. Route by visual control:
- text plus compatible driving audio →
beatra.videos.generate; - strict approved cover art or portrait opening →
beatra.videos.animatewith driving audio; - strict approved opening and ending art →
beatra.videos.interpolate; - audio used only as loose mood reference →
beatra.videos.generate_from_references.
After a delivered gift cover, prefer beatra.videos.animate with driving audio when that still is the opening frame. Inspect the real audio duration and write the smallest admitted whole-second video duration at or above that length. Do not shorten the track to cheapen the clip. If no current card admits a duration that contains the excerpt in full, stop and ask for the smallest compatible excerpt change, or, with the music tools, offer a trim to that length and name the part kept.
For a song Beatra generated, offer a song video (MP4) as another option beside the generated clip, comparing the two live prices, when this account has the music tools: beatra.music.edit with operation: "deliverable" and kind: "video_mp4" makes a song video over its cover art, from the song as generated (not from the cut excerpt). Its video_options.author and video_options.brand_text carry only words the user gave. It needs the song exactly as Beatra generated it; a trimmed or uploaded song cannot take it. Its own card shows the work, the live price from beatra.models.list (capability music_video), one call, its own client_request_id, that the song stays usable if the user stops, and that a short balance is relayed with the returned message and its top-up URL exactly, with a retry of the same ID only after the user says they have topped up; the 600-credit line and the top-up confirmation belong only to the video clip's admission card.
Golden path
-
Inspect the music excerpt and every available image. Record the audio's actual MIME type, byte size, and real duration. Record each image's actual MIME type, width, height, aspect ratio, and byte size. Identify the hook, energy, dominant mood, pulse, and the visual direction this clip should deliver.
-
Build an audio-visual map: hook or energy, dominant mood, pulse, visual motif, subject, movement, palette, camera, opening image, and landing image. Then write one positive prompt that encodes the visual response to the music for this single clip.
-
If no usable cover or opening still exists, call
beatra.models.listwith{"capability":"text_to_image"}, show that image stage's own card, and submit onebeatra.images.generatecover or keyframe. Deliver and inspect that still before any video work. This gift-sized stage does not authorizebeatra.videos.generate,beatra.videos.animate,beatra.videos.interpolate, orbeatra.videos.generate_from_references. For example:printf '%s' '{"prompt":"Album-cover keyframe of a neon-soaked night drive through rain, cinematic teal-and-magenta palette.","count":1,"client_request_id":"opaque-mv-cover-id"}' | python3 scripts/mcp_client.py call beatra.images.generate -
Select the route by visual control. Call
beatra.models.listwith the matching capability to confirm a current card admits every actual media fact, the driving-audio combination, and the selected video duration. Keep the model atautounless the user chose a concrete eligible model. Write the smallest admitted whole-seconddurationat or above the real song length. Do not omitdurationonmodel=auto. Any numeric estimate is provisional; the terminal task'sbilling.net_charged_creditsis final. -
Show an admission card before any video
client_request_idorbeatra.videos.generate/beatra.videos.animate/beatra.videos.interpolate/beatra.videos.generate_from_referencescall: route and MCP tool name, live-card duration, resolution if set, aspect if set, provisional live estimate, the fact that the 600-credit signup gift usually cannot start this video, and what happens if the balance is short. Planning, comparison, or “make the clip” is not approval. Do not submit until the user confirms they have topped up or already have enough credits for this estimate. -
Freeze the exact route, audio, source images, reference media when used, prompt, selected duration, model behavior, and one opaque stable
client_request_id; then submit exactly one video call. Invoke only the bundledscripts/mcp_client.py: the MCP tool name is the CLI argument and its arguments are JSON on standard input. For example, text plus driving audio:printf '%s' '{"prompt":"A neon-soaked night drive through a rain-streaked city, reflections shimmering on the windshield, the beat syncing with passing streetlights, cinematic teal-and-magenta palette.","audio":{"type":"artifact","artifact_id":"art_song"},"duration":10,"client_request_id":"opaque-mv-generate-id"}' | python3 scripts/mcp_client.py call beatra.videos.generateStrict cover art or portrait with driving audio:
printf '%s' '{"image":{"type":"artifact","artifact_id":"art_cover"},"prompt":"The cover art comes alive: light blooms across the frame, particles drift upward, the subject's eyes slowly open in time with the music.","driving_audio":{"type":"artifact","artifact_id":"art_song"},"duration":8,"client_request_id":"opaque-mv-animate-id"}' | python3 scripts/mcp_client.py call beatra.videos.animateStrict opening and ending art:
printf '%s' '{"first_frame":{"type":"artifact","artifact_id":"art_open"},"last_frame":{"type":"artifact","artifact_id":"art_end"},"prompt":"A surreal transition from a desolate moonlit shore to a vibrant sunrise festival, the energy building with the music.","driving_audio":{"type":"artifact","artifact_id":"art_song"},"duration":10,"client_request_id":"opaque-mv-interpolate-id"}' | python3 scripts/mcp_client.py call beatra.videos.interpolateAudio as loose mood reference:
printf '%s' '{"references":[{"kind":"audio","media":{"type":"artifact","artifact_id":"art_song"}},{"kind":"image","media":{"type":"artifact","artifact_id":"art_mood"}}],"prompt":"A dreamlike sequence of a dancer moving through shifting colored smoke, the visuals guided by the song's energy and the mood reference palette.","duration":10,"client_request_id":"opaque-mv-references-id"}' | python3 scripts/mcp_client.py call beatra.videos.generate_from_referencesDo not configure, call, or use a host Beatra Connector. Do not use REST/OpenAPI fallback. Submit the chosen video tool exactly once.
-
Record the returned task ID immediately and poll the same task with
beatra.tasks.getuntil terminal. Deliver every returned video artifact or link. Report only actual returned task status, resolved model, dimensions, duration, usage, andbilling.net_charged_credits. Review accessible output for visual response to the music, motion, mood, subject stability, presence and handling of the track, ratio, and actual duration. State what the host Agent could and could not inspect.
Paid changes, recovery, and cancellation
The section cut, the optional cover stage, the video stage, and the song video (MP4) are separate paid requests with distinct IDs. A changed music excerpt, visual direction, cover image, prompt, model, selected duration, aspect ratio, or video control is new logical paid work: create a new ID and show that stage's changed card; for the video clip, that is the admission card, with fresh top-up or balance confirmation. Never reuse an ID across changed arguments. On insufficient_balance, relay the returned message, keep the top-up URL inside the balance error exact, and retry the same frozen ID only after the user says they have topped up.
If a create response is lost, retry only the identical frozen payload with the same stage ID. If a task ID is lost, call beatra.tasks.list for the relevant capability, inspect plausible candidates with beatra.tasks.get, and match them against that stage's private ledger before considering an identical retry. Queued and running are progress states, not failures. Recover the original stage before planning changed work; never duplicate a paid submission or guess its charge or refund.
Call beatra.tasks.cancel only when the user asks to cancel. Call it once and confirm the resulting terminal state with beatra.tasks.get. A 409 means cancellation is not confirmed, so continue polling that same task without creating replacement work.
If the tool returns insufficient credits: give the user the tool message in full. Do not change the URL. You may translate the other words. Do not replace it with your own top-up copy. Point the user to pay using that message.
Account balance
When the user asks how many credits remain or whether a live estimate fits,
call beatra.wallet.get. When they ask what was charged, call
beatra.wallet.ledger. Both are read-only. Do not invent an account-balance or
top-up tool. Do not make wallet.get a required step before every paid submit.
When a model card comes back carrying a top_up block, relay its tiers as the
card lists them and in that order. Do not rank them, do not talk one down, and
do not pick one for the user. Which tier suits them is their call, made on
the wallet page with the whole list in front of them. Never quote a tier from
memory.
References by task
- Read Music video clip workflow when building an audio-visual map, choosing a route, checking live model facts, constructing exact payloads, polling, recovering, cancelling, or reviewing the visual response to the music.
- Read Music editing and audio when the song needs cutting to the clip's section or the user wants a song video (MP4) of a Beatra song.
- Read Installation and authentication only when authorization or shared credentials need attention.
- Read Installation registration for the non-billable best-effort package registration step.
- Read Tasks and results for shared terminal task and artifact semantics, and Billing, errors, and recovery for returned billing or error details.
- Read Bundled MCP Client diagnostics when the bundled client cannot connect. Do not configure a host Connector.
- Updates and security: updates and security
- Read Uninstall and disconnect only when the user asks to remove the package or shared credentials.
Use or invoke the bundled scripts/mcp_client.py for every Beatra operation.
Updates and security
This Skill runs only the files installed with it and leaves them unchanged while it runs; it never downloads or executes other code. It talks to Beatra's official services over HTTPS, and a file you choose to upload goes only to the one-time upload address Beatra issues for it.
Once a day at most, the bundled client reads Beatra's public version list. When the store or tool you installed from has a newer version, it prints one line with the command that updates it there. Updating stays your choice, through that store or tool. See updates and security.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |