Getting Started

Drop your song lyrics, genre, or track mood and I'll generate a full music video concept with scenes, shot ideas, and visual direction. No lyrics? Just describe the feeling you want your video to have.

Try saying:

• "I have an indie pop song about heartbreak with a slow tempo — can you write a storyboard with 8 scenes that match the emotional arc of the lyrics?"
• "Create a music video concept for a high-energy hip-hop track. I want street scenes, quick cuts, and a color grade that feels gritty but cinematic."
• "My band is shooting a music video this weekend with minimal budget. Give me a shot list and location ideas for a melancholic acoustic folk song set outdoors."

First-Time Connection

When a user first opens this skill, connect to the processing backend automatically. Briefly let them know (e.g. "Setting up...").

Authentication: Check if NEMO_TOKEN is set in the environment. If it is, skip to step 2.

1. Obtain a free token: Generate a random UUID as client identifier. POST to https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token with header X-Client-Id set to that UUID. The response data.token is your NEMO_TOKEN — 100 free credits, valid 7 days.
2. Create a session: POST to https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent with Authorization: Bearer \x3Ctoken>, Content-Type: application/json, and body {"task_name":"project","language":"\x3Cdetected>"}. Store the returned session_id for all subsequent requests.

Keep setup communication brief. Don't display raw API responses or token values to the user.

From Beat Drop to Final Cut — Visually

Music videos are no longer just for major label artists with six-figure budgets. Music Video Maker AI gives independent musicians, YouTubers, and creative directors the tools to conceptualize, script, and sequence compelling music videos — all driven by the mood, tempo, and lyrics of their track.

This skill analyzes the emotional arc of your song and translates it into visual language. Describe your track's genre, vibe, or paste your lyrics, and you'll get scene-by-scene storyboard ideas, shot type suggestions, color palette recommendations, and transition cues timed to your music's structure. Think of it as having a creative director on call.

Whether you're planning a DIY shoot with a smartphone or directing a full production, Music Video Maker AI helps you arrive on set with a clear vision. No more blank-page paralysis — just a focused creative brief that makes every second of your video intentional and impactful.

Routing Your Visual Requests

Every prompt you send — whether it's a mood, genre, lyric snippet, or full creative brief — gets parsed and dispatched to the appropriate generation pipeline based on video style, beat-sync requirements, and visual complexity.

Cloud Rendering API Reference

Music Video Maker AI runs on a distributed cloud rendering backend that processes audio waveform analysis, scene segmentation, and frame generation concurrently to deliver beat-matched visuals at scale. Render jobs are queued, prioritized by resolution tier, and returned as streamable video assets once the pipeline completes synthesis.

Skill attribution — read from this file's YAML frontmatter at runtime:

• X-Skill-Source: music-video-maker-ai
• X-Skill-Version: from frontmatter version
• X-Skill-Platform: detect from install path (~/.clawhub/ → clawhub, ~/.cursor/skills/ → cursor, else unknown)

All requests must include: Authorization: Bearer \x3CNEMO_TOKEN>, X-Skill-Source, X-Skill-Version, X-Skill-Platform. Missing attribution headers will cause export to fail with 402.

API base: https://mega-api-prod.nemovideo.ai

Create session: POST /api/tasks/me/with-session/nemo_agent — body {"task_name":"project","language":"\x3Clang>"} — returns task_id, session_id.

Send message (SSE): POST /run_sse — body {"app_name":"nemo_agent","user_id":"me","session_id":"\x3Csid>","new_message":{"parts":[{"text":"\x3Cmsg>"}]}} with Accept: text/event-stream. Max timeout: 15 minutes.

Upload: POST /api/upload-video/nemo_agent/me/\x3Csid> — file: multipart -F "files=@/path", or URL: {"urls":["\x3Curl>"],"source_type":"url"}

Credits: GET /api/credits/balance/simple — returns available, frozen, total

Session state: GET /api/state/nemo_agent/me/\x3Csid>/latest — key fields: data.state.draft, data.state.video_infos, data.state.generated_media

Export (free, no credits): POST /api/render/proxy/lambda — body {"id":"render_\x3Cts>","sessionId":"\x3Csid>","draft":\x3Cjson>,"output":{"format":"mp4","quality":"high"}}. Poll GET /api/render/proxy/lambda/\x3Cid> every 30s until status = completed. Download URL at output.url.

Supported formats: mp4, mov, avi, webm, mkv, jpg, png, gif, webp, mp3, wav, m4a, aac.

SSE Event Handling

~30% of editing operations return no text in the SSE stream. When this happens: poll session state to verify the edit was applied, then summarize changes to the user.

Backend Response Translation

The backend assumes a GUI exists. Translate these into API actions:

Draft field mapping: t=tracks, tt=track type (0=video, 1=audio, 7=text), sg=segments, d=duration(ms), m=metadata.

Timeline (3 tracks): 1. Video: city timelapse (0-10s) 2. BGM: Lo-fi (0-10s, 35%) 3. Title: "Urban Dreams" (0-3s)

Error Handling

Best Practices

Before diving into full storyboard generation, start with a one-paragraph brief about your artist identity, target audience, and the story you want the video to tell. This context shapes every creative decision the AI makes — from casting suggestions to set design ideas.

Always request scene descriptions that include shot type (wide, close-up, tracking), lighting mood, and action happening on screen. Vague outputs are usually the result of vague inputs. The more cinematic language you use in your prompt, the more cinematic the response.

If you're working with a real production crew, ask Music Video Maker AI to format the output as a proper shot list or call sheet-style document. This makes it immediately usable on set and bridges the gap between AI-generated concepts and real-world execution. Revisit the concept after your first rough cut to get re-edit and pacing suggestions.

Troubleshooting

If your generated concept feels too generic, it usually means the prompt lacked specificity. Avoid broad descriptors like 'cool' or 'modern' — instead use references like 'early 2000s MTV aesthetic' or 'A24 film color grading' to anchor the creative direction.

Getting scenes that don't match your song's energy? Try explicitly stating the BPM or describing the song structure (e.g., 'verse is slow and whispered, chorus explodes with drums'). Music Video Maker AI responds well to structural cues that mirror how a real director would read a track before shooting.

If the storyboard feels too long or too short for your song's runtime, specify the exact duration and number of scenes you need. For a 3-minute song, asking for 12 scenes gives roughly 15 seconds per scene — a natural rhythm for most music videos. You can always ask for scene expansions or cuts to tighten the pacing.

Tips and Tricks

Get the most out of Music Video Maker AI by giving it as much context as possible upfront. The more specific you are about your song's tempo, genre, lyrical themes, and intended audience, the more tailored and production-ready your output will be.

Try pasting your full lyrics directly into the prompt — the AI uses line breaks, emotional peaks, and recurring hooks to time scene transitions and build a narrative structure that mirrors your song's flow. Chorus sections often translate best into high-energy visual moments or repeated motifs.

Don't overlook the power of color and mood prompts. Phrases like 'neon-drenched nighttime' or 'sun-bleached desert road' give the AI a cinematic anchor to build around. You can also ask for multiple concept variations and cherry-pick elements from each to create a hybrid vision that's uniquely yours.