levea-mcp-server

Levea MCP Server

An autonomous agentic video editor. Send a sentence. Get a finished video. No timelines, no keyframes, no plugin chains. Open-source MCP client connecting local AI agents (Claude, Cursor, Cline) to our cloud GPU video production harness.

Beta. The agent can make mistakes. Preview every output before publishing or sharing. For high-stakes or irreversible edits, pass requirePlanApproval: true so the agent stops after planning and waits for your approval before anything runs.

---

Contents

1. Quickstart — running in 60 seconds
2. What it does — say it, get it
3. Capability surface — the full toolset
4. Goals & briefs — from one line to a full creative brief
5. API reference — endpoints, request/response, streaming, async
6. Who it's for · vs. legacy editors · Safety · Media
7. Links & channels

---

Quickstart

The portable interface is the levea-mcp-server MCP server — one server, every MCP client (Claude Desktop, Claude Code, Cursor, Cline, OpenClaw, Hermes), one tool surface, one backend contract, so nothing drifts per platform.

1. Get an API key — sign up at livecore.ai and generate an OpenClaw API key.

2. Add the MCP server — the same npx line works for every MCP client:

{
  "mcpServers": {
    "levea": {
      "command": "npx",
      "args": ["-y", "levea-mcp-server"],
      "env": {
        "LEVEA_API_URL": "https://api.livecore.ai",
        "LEVEA_API_KEY": "your-key-from-livecore.ai"
      }
    }
  }
}

3. Use it — say to your AI: "Use the levea tool to generate 5 viral clips from this video, add captions, reframe vertical." The agent picks the tool, fills the prompt; the editor plans → executes → exports.

Per-client setup

Client	How
Claude Desktop / Cursor / Cline	Add the levea block above to mcpServers — see the levea-mcp-server docs
Claude Code	claude mcp add levea -e LEVEA_API_URL=https://api.livecore.ai -e LEVEA_API_KEY=... -- npx -y levea-mcp-server
OpenClaw	This ClawHub listing (chat-native) or openclaw mcp add levea --command "npx -y levea-mcp-server" --env LEVEA_API_URL=https://api.livecore.ai --env LEVEA_API_KEY=...
Hermes	Register levea-mcp-server as an MCP server in your Hermes config

This package on npm: npx -y levea-mcp-server — no global install needed.

Hosted endpoint — no install

Clients that connect to a remote MCP server by URL (Smithery, Claude / ChatGPT connectors, Cursor remote servers) can skip npx entirely:

URL:    https://api.livecore.ai/api/v1/misc/mcp      (Streamable HTTP, stateless)
Header: Authorization: Bearer your-key-from-livecore.ai

Same tools, same backend. Clients that can't set headers may pass ?apiKey=... instead. Tool listing works without a key; edits need one.

Every client calls the same backend — the MCP server, this ClawHub listing, and every other surface share one tool contract.

Two hosts — don't confuse them

Host	Use
https://api.livecore.ai	Functional API. Set as LEVEA_API_URL. The client appends /api/v1/misc/openclaw/v1/execute automatically — never put a full path here.
https://livecore.ai	API-key portal (UI). Get your key here. Do not use as LEVEA_API_URL.

Env vars are LEVEA_API_URL / LEVEA_API_KEY.

Tools the AI sees

One editing entry point. Everything else is typed state-management and polling — so integrators can build full editing experiences through one API key instead of mixing JWT and MCP.

Group	Tools
Edit (one entry point, 3 variants)	autonomous_edit (primary, JSON) · autonomous_edit_streaming (SSE — per-step progress) · queue_edit (async fire-and-forget)
Job / task polling	check_job_status · check_task_status · get_active_task
Caption templates	list_caption_templates · apply_caption_template · save_caption_template · save_current_caption_template · delete_caption_template (41 builtin + your saved)
Brand kits	list_brand_kits · get_brand_kit · create_brand_kit · update_brand_kit · delete_brand_kit (palette · fonts · logo · voice · gradeBias · enforcement)
Projects	list_projects · get_project · create_project · delete_project
Assets	asset_upload_url (signed PUT) · list_assets · delete_asset · transcribe_asset
Diagnostics	editor_health

No edit shims (add_captions, generate_viral_clips, …) and no editor_execute escape hatch. The backend's planner is the specialist — fragmenting the editing surface encourages calling LLMs to second-guess the planner and lose multi-step intent. State-management tools (brand / project / asset / caption-template CRUD) aren't editing, so they don't fragment intent — they just give integrators typed access.

→ For agent-integration details (schemas, async semantics, error handling, plan approval), see AGENTS.md.

---

What it does

Say it in plain language; the agent plans and finishes the edit.

You ask	It does
"Turn this into 5 viral clips with captions and vertical reframe"	Picks the best moments, cuts, captions, reframes, exports
"Cut a 60-second highlight from this 2-hour podcast"	Identifies narrative peaks, trims, packages with captions
"Make this TikTok-ready"	Vertical 9:16 reframe + captions + silence removal + emphasis kit
"Export for TikTok, Reels, Shorts, YouTube, and Instagram"	One pass, all aspect ratios in one bundle
"Replace the green screen with a beach, keep the speaker centered"	Chroma key + background composite + motion tracking in one call
"Remove the background — no green screen"	AI background removal via alpha matte (any footage)
"Swap my background for a city skyline — no green screen"	AI matte removes the subject cleanly, then composites the new background behind
"Remove all silences and filler words, add background music"	Cleans the audio track, adds ducked music under speech
"Auto-zoom on whoever's talking"	Active-speaker detection + dynamic zoom-follow framing
"Caption this and highlight every time they say 'launch'"	Auto-captions + keyword emphasis (scaling / glow / pulse)
"Find every clip where Alex appears"	Cross-asset face identity search
"Add narration in a cloned voice over the intro"	Voice cloning + TTS overlay + auto-ducking
"Caption this Hormozi-style with karaoke word highlighting"	Word-synced karaoke captions — active word fills + underlines in time with speech
"Generate B-roll over the product mention"	AI B-roll generation + placement at the right timestamp
"Color grade this like a Netflix doc"	Color grading with cinematic preset
"Slow-mo the climax, freeze on the reveal"	Speed change (slow-mo / fast-forward) + freeze-frame
"Reframe to vertical but don't crop the lower-third captions"	Caption-safe 9:16 reframe (detects on-screen text regions and reframes around them)
"Pull the key stats from this and animate them as charts"	Transcript-driven charts + stat-callout motion graphics
"Sync these 3 camera angles and cut between them on the active speaker"	Multi-cam sync + automatic angle switching
"Blur the license plates and bleep the swearing"	Privacy redaction + profanity cleanup

---

Capability surface

The full toolset behind the prompt. Everything below is reachable from natural language — mix and match in a single sentence.

🔍 Read & inspect

• Scene state — timeline / layer / scene structure, schema introspection
• Search — asset gallery (type, duration, name); transcript (keyword, semantic, timestamp)
• Vision — CV frame analysis (object, face, scene); on-screen text-region detection (locates chyrons / lower-thirds / burned-in captions so reframing avoids covering them)
• Intelligence — narrative peaks, speaker diarization, sentiment, pacing
• Jobs — async job status polling

✂️ Structural editing

• Layers — insert / update / replace / delete (video, audio, image, text, shape, solid, adjustment, group, light, vfx, visualizer, lottie)
• Timing — trim, split, retime (slow-mo 0.5×, fast-forward 2×, freeze-frame), reverse, stabilize
• Arrange — reposition, sequence, snap to transcript word boundaries
• Assemble — multi-cam sync + automatic angle switching, montage assembly (shot discovery + sequencing from source footage), slideshow generation, chapter markers, SRT import
• Safety — gap healing, audio normalize, duration reconcile (pre-export pass), multi-step undo / redo

🎬 Visual editing

• Color grade — brightness, contrast, saturation, hue, lift/gamma/gain, RGB curves, LUT / film-look presets, vignette, chromatic aberration
• VFX shaders — smoke, dust, fire, explosion, lightning, snow, glitch, scanlines, grain, glassmorphism, bokeh, lava, corrosion, portal
• Composite — chroma key, AI background removal (alpha matte), masking (luma / alpha / depth, with boolean multi-mask)
• Blend & opacity — 17 blend modes (multiply, screen, overlay, color dodge / burn, difference, hue / saturation / color / luminosity, additive) + per-layer & group opacity
• Shape & transform — clip shapes (circle, dome, star, hexagon, polygon), crop, 3D rotation / perspective, glow, shadow, gradient fills
• Lighting — up to 8 scene lights illuminating 3D-transformed layers
• Reframe & layout — caption-safe vertical reframe (9:16) + vertical montage, split screen (top/bottom, L/R, PiP, grid)
• Track — motion / face tracking with zoom-follow, branding overlays

📊 Motion graphics & data viz

30+ production-grade animated graphics:

• Data — transcript-driven charts (bar, line, pie), stat callouts, animated counters
• Lower-thirds & cards — lower-thirds, title / chapter cards
• Diagrams — flowcharts, timelines, geo callouts
• CTAs — animated subscribe / call-to-action buttons

💬 Captions & text

• Generate — auto-captions from the transcript
• Style — built-in templates, or an AI director that picks/generates one at runtime
• Animate — per-word animations (typewriter, slide, fade, scale, rotate, bounce, flip, swing, elastic, blur, glitch, wave — each with a matching exit)
• Typography — crisp scalable glyphs (MSDF), stroke / outline, drop shadow, gradient fill, background pill / box, per-word karaoke fill + underline
• Shape — curved text paths (circle, wave, custom SVG), Lottie playback control

🎵 Audio

• Clean up — remove silences, breaths, filler words; word-level mute/cut
• Mix — auto-duck (sidechain music vs voice), loudness (LUFS) normalize + true-peak limiting, denoise, crossfade, EQ presets (bass-boost, vocal-clarity, warm, bright), stem separation
• Generate — SFX, music (mood/genre/BPM), voiceover (TTS or cloned voice)
• Sync — external master audio alignment (timecode / explicit offset)
• Visualize — waveforms (bars, wave, circular)

🔒 Privacy & cleanup

• Faces — blur all faces or background-only
• Regions — privacy redaction (blur / cover)
• Language — profanity cleanup (mute / bleep)
• Layout — safe-zone repair

✨ Async generation

• Video — AI video / B-roll (duration + aspect)
• Images — AI images (single or batch at timestamps), generative image edit, auto-thumbnail extraction
• Audio — AI music (prompt + duration + mood + genre + BPM), AI voiceover (TTS or cloned voice library)

🚀 High-level presets

• Viral — vertical + captions + silence removal + tracking + word emphasis
• Cinematic director — dynamic zooms + cinematic grade + mood camera moves
• Emphasis system — keyword detection + coordinated scaling / glow / pulse
• Pacing optimizer — filler + silence + low-energy removal for retention

📦 Export

• MP4 — single file (resolution / codec / quality tier)
• Viral-clip batch — auto-segmented short-form, packaged as a ZIP
• Multi-platform pack — TikTok + Reels + Shorts + YouTube + Instagram in one pass

✅ Also shipped

• Content-aware inpainting (LaMa) — removes on-screen text, logos, and watermarks via REPLACE_ON_SCREEN_TEXT and overlay-removal powers
• On-screen text OCR recognition — PP-OCRv5 detection + recognition populate TextRegion.text, enabling find-and-replace on burned-in text

---

Goals & briefs

autonomous_edit accepts anything from a five-word command to a five-hundred-word creative brief.

1. Direct commands — a single edit, single result. The What it does examples above are all direct commands.

2. Open-ended goals — hand the agent a goal instead of a command; it inspects the asset, then proposes a plan:

"Watch this and tell me how to make it more engaging for TikTok"
"Look at the first 30 seconds and suggest 3 ways to hook the viewer"
"Review this footage and propose edits to tighten the pacing"

Pair these with requirePlanApproval: true for propose-only mode — the agent stops after planning and waits for your approval (see Plan approval).

3. Full creative briefs — multi-step orchestrations executed end-to-end. The agent decomposes the brief into a DAG of canonical actions, plans the order, executes through safety gates, verifies each step, and exports every requested format. Use SSE to watch each step land in real time.

Example brief (~20 actions composed in one planned run)

Transform this video into a viral social media documentary. Start by trimming the first 3 seconds and last 5 seconds to remove dead air. Apply a cinematic color grade with warm tones and high contrast. Generate auto-captions with bold yellow text and black outline, positioned bottom center. Add a dynamic zoom-in on the main subject during the first 10 seconds. Insert 3 contextual b-roll clips where the speaker mentions visual concepts. Apply face tracking to keep the subject centered when converting to vertical 9:16 for TikTok and Reels. Add an upbeat background track that matches the energy, with auto-ducking when the speaker talks. Remove all silence gaps longer than 0.5 seconds. Apply a subtle vignette. Add my logo watermark top-right at 80% opacity. Generate a custom thumbnail from the most expressive frame with bold "MUST WATCH" text. Create an animated subscribe button that appears at 5 seconds and pulses. Add whoosh SFX for transitions. Apply noise reduction. Split at 30 seconds to insert a 2-second transition. Add animated counter overlays for key statistics. Blur background faces for privacy. Export in 1080p for YouTube plus optimized TikTok, Reels, and Shorts versions with platform-specific aspect ratios.

No timeline UI, no per-step API calls, no glue code on your side.

---

API reference

Base URL: {LEVEA_API_URL} (production: https://api.livecore.ai). Auth on every request:

Authorization: Bearer {LEVEA_API_KEY}

All paths are under /api/v1/misc/openclaw.

Endpoint	Purpose
POST /v1/execute	Run autonomous_edit (the main entry point)
POST /v1/queue-edit	Fire-and-forget async edit
GET /v1/jobs/{jobId} · GET /v1/task-status/{taskId}	Poll async render / generation jobs
GET /v1/projects/{id}/active-task	The project's in-flight task, if any
GET /v1/tools · GET /v1/health	List tool names · health check
/v1/brands · /v1/projects · /v1/assets/*	Brand-kit / project / asset CRUD — list · get · create · update · delete
/v1/caption-templates (+ /apply, /save-current)	Caption-template CRUD + apply

Each management group above is also exposed as a typed MCP tool (see Tools the AI sees).

LEVEA_API_URL is the bare host https://api.livecore.ai — not a full path, not livecore.ai, and not the in-product /api/v1/misc/editor/ route.

Request

Every edit goes through one tool — autonomous_edit. Pass a natural-language description; the agent plans, executes, verifies, and exports. No tool list to memorize, no structured params to learn.

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

prompt is the only required field. Everything else is optional:

Field	What it does
prompt	Description of the video edit to perform, in the user's own words. (required)
project_id	Project identifier returned by a prior call; reuse it for every follow-up edit on the same video.
video_url	Public source video URL when starting a new project.
assets	Input media (video, image, audio) as public HTTP/HTTPS URLs.
scene	Existing scene/timeline state to continue editing (advanced).
captionTemplatePreset	Caption template id to apply: a builtin slug (e.g. hormozi, viral-pop, karaoke) or a saved template id. Call list_caption_templates for the exact ids.
captionTemplateMode	How to apply captionTemplatePreset: one of add, recreate, style, translate, clear, remove_fillers.
requirePlanApproval	When true, pause after planning and return workingMemory for the user to approve.
workingMemory	State returned by a paused plan-approval call; send it back to resume.
editedPlan	The user's edited version of a plan returned by an approval pause.
attachedImages	Reference image URLs for planning (style, branding).
flaggedIssues	Specific problems the user flagged that this run must fix.
brandId	Brand kit to apply to this call.
projectBrandId	Default brand kit for the project; persists across calls.
core_only	Return only rendering-essential scene fields (smaller responses).

Response

{
  "type": "success | partial_success",
  "tool": "autonomous_edit",
  "success": true,
  "partial": false,
  "status": "completed | failed | awaiting_approval",
  "scene": { },
  "reply": "Human-readable summary of what changed",
  "message": "Short status message",
  "videoUrl": "https://.../output.mp4",
  "jobId": "12345",
  "viral_clips": [ ],
  "zip_url": "https://.../clips.zip",
  "activeTasks": [ ],
  "pendingAsyncJobs": [ ],
  "verificationPassed": true,
  "verificationIssues": [],
  "quality": {
    "score": 0.94,
    "stepsTotal": 7,
    "stepsFailed": 0,
    "hasExportedMedia": true,
    "mediaReachable": true
  },
  "processingTime": 12.3,
  "workingMemory": { }
}

jobId is the numeric task_id (a string of digits), not a task_-prefixed value. quality is a calibrated post-run score block; partial is the top-level partial-success flag (mirrors type: "partial_success").

SSE streaming

Set Accept: text/event-stream or ?stream=true. Notable event types:

Event	Meaning
heartbeat	15s keepalive
status · phase	Phase / status transitions
thought · reasoning_chunk	Per-step reasoning visibility
tool_call · tool_result	Action started / finished
background_job_completed	Async job done
success / partial_success	Terminal payload
error	Terminal failure

Async job lifecycle

Generation actions return immediately with a jobId. Poll until terminal:

JOB_ID=$(jq -r '.jobId // .activeTasks[0].intent.job_id' /tmp/result.json)
while true; do
  STATUS=$(curl -sS "$LEVEA_API_URL/api/v1/misc/openclaw/v1/jobs/$JOB_ID" \
    -H "Authorization: Bearer $LEVEA_API_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  case "$STATUS" in completed|succeeded|failed|error|cancelled) break;; esac
  sleep 5
done

curl -sS "$LEVEA_API_URL/api/v1/misc/openclaw/v1/jobs/$JOB_ID" \
  -H "Authorization: Bearer $LEVEA_API_KEY" | jq '.result'

Auto-export

After any mutating tool call that changed the scene (including partial successes), if no export is already queued the route auto-fires one synchronously — so the response carries a real videoUrl. Read-only and conversational autonomous_edit calls do not trigger auto-export.

Plan approval

Pass requirePlanApproval: true to make the agent stop after planning. It returns status: "awaiting_approval" + populated workingMemory. Resume by sending the same workingMemory with an approval prompt (yes, approve, go, do it, confirm).

---

Who it's for

Audience	What they get
Creators & influencers	Long videos → TikToks / Reels / Shorts · auto-captions · viral-clip generation · AI thumbnails
Podcasters	Video-podcast highlights · audiograms · multi-cam editing · silence + filler-word removal
Marketers & agencies	Ad creation at scale · social-first repurposing · brand-overlay automation · multi-platform exports
Educators & course creators	Tutorial editing · auto-chapters · title / chapter cards · auto-captions · word-level emphasis
YouTubers	Long-form → Shorts pipeline · thumbnail generation · intro/outro automation · chapter markers
Sales & SaaS	Product demos · walkthrough highlights · cloned-voice narration
Event teams	Webinar / conference highlights · multi-cam stitching · speaker spotlight reels
Faith & community	Sermon clips for social · message highlights · multi-platform packaging

---

vs. legacy editors

	Legacy (Premiere, DaVinci, CapCut, Descript)	Levea MCP Server
Interface	Drag, drop, keyframe by hand	One sentence; the agent does the rest
Auto-analysis on upload	Manual scene detection + subtitles	Faces, speakers, shots, on-screen text regions, mattes detected automatically
"Make this viral"	You build the workflow	Single preset — vertical + captions + silences + tracking
Cross-asset search	Filename search	"Find every clip where Alex appears" across the whole library
Background replace	Key out, find background, composite	One call
Broadcast-grade audio	Manual loudness metering + EQ	Auto LUFS loudness-normalized + true-peak limited, music auto-ducked under speech
Editorial reasoning	You decide the climax	Agent surfaces narrative peaks and reaction moments
Verification	You eyeball it	Verifier runs after execution; auto-repairs on failure
Multi-platform export	One render per ratio	TikTok + Reels + Shorts + YouTube + Instagram in one pass
Extensibility	Plugins call external binaries	Typed tools called by name

This isn't an editor with AI features bolted on — it's an autonomous agent that finishes the edit for you.

---

Safety & limits

• Beta — outputs can be wrong. Preview every result before sharing. For irreversible workflows, pass requirePlanApproval: true to inspect and approve the plan first.
• API-key authentication on every request.
• Destructive actions (mass deletes, clears) require explicit confirmation params.
• Output is verified after execution; auto-repairs on failure.
• Concurrent identical requests 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 media

	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 generation both supported
Recommended resolution	1080p or 4K; canvas configurable per project

---

Links & channels

Channel	Identifier / link
npm (MCP server)	levea-mcp-server
MCP Registry	io.github.brajendrak00068/levea-mcp-server (registry)
ClawHub plugin	openclaw-ai-video-editor
ClawHub skill (agentic)	ai-agentic-video-editor
ClawHub skill (Levea brand)	levea-ai-video-editor
Sign up + API keys	livecore.ai
API base	https://api.livecore.ai

Versions aren't pinned here — npx -y levea-mcp-server and the ClawHub listing always pull the latest, and the linked npm / ClawHub pages show the current version. You never need a specific number.

Support — setup help, integration questions, or issue reports: [email protected]

License — MIT