Levea Agentic Video Editor

A full agentic video-editing surface. Send a prompt and get a planned, verified, executed edit. The autonomous editor chooses its own internal tools, applies safety gates, exports when needed, and returns the final scene/video result. Every edit goes through one HTTP tool — autonomous_edit — so there is only one call to learn.

When to use this skill

Use when the user asks for AI video editing, natural-language video edits, viral clips, TikTok videos, Instagram Reels, YouTube Shorts, auto captions / subtitles, chroma key / green-screen removal, background removal, B-roll, motion tracking, motion graphics, multi-cam editing, smart jump cuts, silence / audio cleanup, voiceover, music generation, object / face blur, privacy redaction, beat sync, brand kits, thumbnails, style presets, vertical (9:16) reframe, safe-zone repair, final delivery checks, export presets, multi-platform export, or MP4 export.

Beta: outputs can be wrong. Before executing any mutating edit on user content, describe the planned edit and ask for explicit confirmation. For destructive or irreversible workflows, pass requirePlanApproval: true so the editor halts after planning and the user can approve before execution.

Setup

This skill calls the Levea HTTP API with curl and parses responses with jq.

Required binaries: curl, jq.

Required environment variables:

• ADSCENE_API_URL — base URL for the Levea API, e.g. https://api.livecore.ai. Do not use the Studio URL or the in-product /api/v1/misc/editor route.
• ADSCENE_API_KEY — API key generated from Studio at https://studio.livecore.ai/ (sign up / sign in, then create a key). Studio is only for signup, login, and key management.

export ADSCENE_API_URL="https://api.livecore.ai"
export ADSCENE_API_KEY="your-api-key"

Endpoint

POST {ADSCENE_API_URL}/api/v1/misc/openclaw/v1/execute

Auth: Authorization: Bearer {ADSCENE_API_KEY}

Accepts single-shot JSON (default) or SSE (Accept: text/event-stream or ?stream=true).

Request body:

{
  "tool": "autonomous_edit",
  "params": { "prompt": "..." },
  "project_id": "optional-project-id",
  "scene": { "_comment": "optional client scene; server-side committed scene wins if newer" }
}

How to use it

Pass a natural-language description in params.prompt. The agent classifies intent, decomposes into atomic steps, plans, executes through safety gates, and verifies the result.

curl -sS -X POST "$ADSCENE_API_URL/api/v1/misc/openclaw/v1/execute" \
  -H "Authorization: Bearer $ADSCENE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "autonomous_edit",
    "params": {
      "prompt": "Make this a TikTok-ready viral clip: vertical reframe, add bold captions, remove silences, and motion-track the speaker."
    },
    "project_id": "my-project"
  }'

Behind a single autonomous_edit call the agent can compose any of:

Read / inspect — timeline + layer structure, CV frame analysis (object/face/scene), transcript search (keyword / semantic / timestamp), video intelligence (narrative peaks, diarization, sentiment, pacing), asset-gallery search, async job polling, property-schema introspection.

Structural editing — insert / update / replace / delete layers (video, audio, image, text, shape, solid, adjustment, group, light, vfx, visualizer, lottie); trim / split / retime (slow-mo, fast-forward, freeze-frame); smart jump cuts, filler-word + silence + low-energy cleanup; timeline sequencing, multi-cam sync + angle switching, gap heal, duration reconcile, multi-step undo / redo.

Visual editing — color grade (brightness/contrast/saturation/hue, lift-gamma-gain, RGB curves); procedural VFX shaders (smoke, dust, fire, lightning, snow, glitch, grain, bokeh, lava, portal, …); chroma key (similarity/smoothness/spill, luma/alpha/depth masks); clip shapes; crop + 3D rotation/perspective; glow / shadow / gradient fills; vertical reframe (9:16) + montage; split screen + PiP; brand overlays / kits; motion + face tracking with zoom-follow; object hide / blur, face blur, privacy redaction, safe-zone repair.

Captions & text — auto captions from transcript; built-in templates or an AI caption director; motion graphics (kinetic captions, lower thirds, stat callouts, charts, comparison overlays, concept-icon Lottie); curved text paths; per-word entrance / exit animations; Lottie playback control.

Audio — silence / breath / filler-word cleanup; profanity mute / bleep / cut; auto-ducking; mix / normalize / denoise / EQ; external master-audio sync; beat-synced cuts (beat_times or bpm); SFX, music generation (mood / genre / BPM), voiceover (TTS or cloned); waveform visualizers.

Async generation — AI video / B-roll, AI images (single or batch at timestamps), AI music, AI voiceover, auto-thumbnail or AI thumbnail variants, face blur, instruction-based image edit.

High-level kits (each one canonical action orchestrating many edits) — APPLY_VIRAL_KIT, APPLY_CINEMATIC_DIRECTOR, APPLY_EMPHASIS_SYSTEM, OPTIMIZE_PACING.

Export — EXPORT_VIDEO (MP4), EXPORT_PRESET (platform / codec / aspect presets + safe-zone repair), FINAL_DELIVERY_CHECK, GENERATE_VIRAL_CLIPS (auto-segment short-form clips as ZIP), GENERATE_MULTI_PLATFORM (TikTok + Reels + Shorts + YouTube + Instagram in one pass).

Auto-export follow-up

After any mutating autonomous_edit call where the scene actually changed and the agent did not already queue an export, the route fires one automatically as a second run. The final response carries videoUrl (when ready) or jobId (for polling). Read-only / conversational calls do not trigger auto-export.

Optional input parameters

Pass any of these inside params (or at the top level of the body):

• prompt — required on every call (the natural-language edit description)
• workingMemory — durable working-memory snapshot; re-send to resume after awaiting_approval
• requirePlanApproval — true stops the agent after planning (awaiting_approval); resume with the same workingMemory + an approval prompt
• attachedImages — array of base64 screenshots / reference images
• flaggedIssues — array of strings describing specific problems to fix
• captionTemplatePreset, captionTemplateMode — caption style routing
• brandId, projectBrandId — brand-kit selection (colors, fonts, logo, grade bias, voice)
• core_only (also ?core_only=true) — minimal scene shape (rendering-only)
• assets — additional asset descriptors to make available to the agent

Response shape (JSON, default)

{
  "type": "success",
  "tool": "autonomous_edit",
  "success": true,
  "status": "completed",
  "scene": { "_comment": "updated scene" },
  "reply": "Human-readable summary of what changed",
  "videoUrl": "https://.../output.mp4",
  "jobId": "1234567",
  "viral_clips": [],
  "zip_url": "https://.../clips.zip",
  "activeTasks": [],
  "pendingAsyncJobs": [],
  "verificationPassed": true,
  "verificationIssues": [],
  "workingMemory": { "_comment": "return this in the next call to resume approval-paused runs" }
}

Failure (HTTP 4xx/5xx): { "success": false, "error": "...", "code": "MISSING_PROMPT" }.

status is one of completed, failed, awaiting_approval.

Async job lifecycle

Generation actions (generate_*, EXPORT_VIDEO) return immediately with a jobId in activeTasks / pendingAsyncJobs. Poll:

curl -sS "$ADSCENE_API_URL/api/v1/misc/openclaw/v1/jobs/$JOB_ID" \
  -H "Authorization: Bearer $ADSCENE_API_KEY" | jq '{status, progress, message, result}'

status is queued | processing | completed | failed. To harvest async-generated content into the timeline, issue an autonomous_edit prompt like "apply any pending generated content".

Plan-approval flow

With requirePlanApproval: true, the response carries status: "awaiting_approval" + a populated workingMemory. To proceed, call again with params.prompt = an approval phrase (yes, approve, go ahead, do it, confirm, …) and the returned workingMemory.

Safety, verification, and limits

Every run flows through three deterministic gates (ActionPermissionGate, ArchitectureControlPlane, EditorSafetyPolicy). Destructive actions (CLEAR, mass deletes) require explicit confirmation params. Verification runs after execution and may trigger up to 2 repair loops; failures surface in verificationPassed: false + verificationIssues[]. Identical concurrent requests for the same (user, project, prompt, scene fingerprint) are deduplicated server-side. Rate-limited per API key. Read-only ~1–3s, structural edits ~3–10s, async generation 30s–5min per artifact, viral-clip / multi-platform exports several minutes.

Supported formats

• Video in: MP4, MOV, WebM (HTTP/HTTPS URLs, YouTube URLs, gallery IDs)
• Image in: JPG, PNG, WebP
• Audio in: MP3, WAV, M4A, AAC (or extracted from video)
• Output: MP4 (export), ZIP (viral clips / multi-platform bundles)
• Max video length: up to ~3 hours per asset (plan-dependent); sync edits and async handles longer
• Recommended resolution: 1080p or 4K; canvas configurable per project

Example: end-to-end viral-clip generation

# 1) Kick off the viral-clip pipeline (auto-export follow-up queues rendering)
curl -sS -X POST "$ADSCENE_API_URL/api/v1/misc/openclaw/v1/execute" \
  -H "Authorization: Bearer $ADSCENE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "autonomous_edit",
    "params": {
      "prompt": "Generate 5 viral clips, 15-30 seconds each, on the most engaging moments. Add bold captions, vertical reframe, remove silences."
    },
    "project_id": "my-project"
  }' | tee /tmp/levea_result.json | jq -r '.jobId // .activeTasks[0].intent.job_id'

# 2) Poll until done
JOB_ID=$(jq -r '.jobId // .activeTasks[0].intent.job_id' /tmp/levea_result.json)
while true; do
  STATUS=$(curl -sS "$ADSCENE_API_URL/api/v1/misc/openclaw/v1/jobs/$JOB_ID" \
    -H "Authorization: Bearer $ADSCENE_API_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  { [ "$STATUS" = "completed" ] || [ "$STATUS" = "failed" ]; } && break
  sleep 5
done

# 3) Fetch the final artifact URL(s)
curl -sS "$ADSCENE_API_URL/api/v1/misc/openclaw/v1/jobs/$JOB_ID" \
  -H "Authorization: Bearer $ADSCENE_API_KEY" | jq '.result'