21 MCP ToolsCloud LibraryEditing EngineAgentic QA Loopsx402 Payments
Generate, refine & manage
media through one endpoint
Composable media generation for AI agents — with a cloud library to manage what they create, an editing engine to refine it, and analysis tools to close the quality loop.
Agents build through MCP. Humans review in the web workspace. Teams collaborate with shared projects. x402 users authenticate with just a wallet.
https://mcp.soundside.ai/mcp·300 free credits · No card requiredIterative Agentic Design
Agents that critique their own work
analyze_media with Gemini vision QA lets agents evaluate their output against an intent checklist, then decide whether to re-generate, edit in post, or approve.
1
Generate
Agent creates image, video, or audio from a prompt
2
Analyze
analyze_media evaluates quality, checks intent, scores output
3
Decide
Agent compares result against spec — pass or iterate?
4
Refine
Re-generate with adjusted prompt, or edit_video to fix in post
5
Store
Approved asset saved to library with lineage and tags
analyze_media(resource_id, analysis_type="vision_qa", intent_checklist={"no_pillarboxing": true})Spec-driven QA: each checklist item becomes an explicit pass/fail check
Collaborative by Design
Three ways in. One shared workspace.
Agents connect via MCP. Humans work in the web UI with the Crew chatbot. x402 users authenticate with a crypto wallet — no account needed. All three share the same projects, assets, and permissions.
🤖MCP Agents
→ JSON-RPC 2.0 endpoint
→ 21 MCP tools
→ Async with status recovery
→ Resource ID chaining
→ Works with Claude, Cursor, any MCP client
💬Web + Crew Chat
→ Project dashboard with media grid
→ Type filters & collections
→ Crew chatbot with full tool access
→ Review agent output & iterate
→ Brand kit management
⚡x402 Wallet
→ USDC on Base via Coinbase
→ No email, no password, no account
→ Pay per call at the endpoint
→ Wallet-based login to web UI
→ Same projects & permissions
lib_share grants view, edit, or admin access by email — enabling team workflows where agents produce, humans review, and collaborators contribute.
Tool Catalog
21 composable MCP tools
6 generation, 6 editing and analysis, 3 library management. Every tool returns a resource ID that chains into any other.
🔍Analyze
analyze_media
Analyze media: technical metadata, vision_qa, transcribe, detect_segments, detect_shots, export_edl.
anthropic · grok · openai · qwen · soundside.ai · vertex
✨Edit
apply_effect
Cinematic effects: ken_burns (image→video motion), speed_ramp, film_grain, vignette. Returns resource_id.
soundside.ai
🧩Edit
compose_media
Layer text, images, or video onto a base video: add_text, overlay, split_screen. Returns resource_id.
soundside.ai
🎞Compose
compose_video
Compose a complete video from a timeline plan.
grok · lyria · minimax
📊Generate
create_artifact
Create a business artifact: presentation (slides), chart (data viz), document (sections), or diagram (mermaid). Bundle mode: pass outputs list for multi-format generation. Supports brand kits. Provider 'gamma' for AI presentations.
docx · gamma · mermaid · plotly · pptx · weasyprint
🔊Generate
create_audio
Create audio: TTS, sound effects, voice cloning/design, voice listing, or the deprecated transcription compatibility shim. Use analyze_media(analysis_type='transcribe') for new transcription integrations.
creative_freedom · grok · minimax · runway · vertex
🖼Generate
create_image
Create image from text prompt or reference. Sync (vertex, minimax, grok, creative_freedom) or async (alibaba — poll with lib_list).
alibaba · creative_freedom · grok · minimax · vertex
🎵Generate
create_music
Create music from lyrics/prompt. Lyria is synchronous; Creative Freedom is authenticated-only and asynchronous. MiniMax music is unavailable and is not part of the public contract.
creative_freedom · lyria
📝Generate
create_text
Generate text via LLM chat completions. Pass messages or prompt.
grok · minimax · qwen · vertex
🎬Generate
create_video
Create video from text/image. Async: returns pending resource_id, poll with lib_list until completed (5-10s intervals, 600s timeout). Supports character_reference for consistency.
alibaba · creative_freedom · grok · minimax · vertex
🎚Edit
edit_audio
Edit audio on video: mix_audio, replace_audio, pad_audio. Returns resource_id.
soundside.ai
✂️Edit
edit_video
Edit video: trim, concat, crossfade, adjust_speed, loop, color_grade, custom, burn_subtitles. Returns resource_id of new media.
soundside.ai
🪄Edit
extract_media
Extract from media: extract_frame, extract_frames, extract_audio. Returns resource_id(s).
soundside.ai
📂Library
lib_list
List library entities: projects, collections, resources, lineage, brand_kits, credits, or dashboard (composite load). Filter by project_id, collection_id, tags, mime_type_prefix. Supports pagination (limit/offset), sorting, and search.
soundside.ai
⚙️Library
lib_manage
CRUD for library entities (projects, collections, resources, brand_kits). Large file upload: upload_initiate → PUT to signed URL → upload_complete. To share a resource: set visibility="public" → response includes a permanent short_url. For listing, use lib_list.
soundside.ai
🔗Library
lib_share
Manage project sharing: share by email, list shares, or revoke access. For download URLs, use lib_list with resource_id.
soundside.ai
📋Adapters
list_adapters
List LoRA adapters mirrored into the Soundside library.
soundside.ai
🛠Adapters
manage_adapter
Manage a LoRA adapter: inspect, deploy, undeploy, delete, or select_checkpoint.
soundside.ai
•Publish
publish_content
Read and manage a connected X account. action defaults to publish. Read actions return bounded posts/profiles and pagination tokens synchronously. Writes (publish/reply/edit/delete/like/unlike/repost/unrepost) require a unique idempotency_key and return a durable receipt: check its status with lib_list; a push may also arrive, but don't rely on it. Reuse a key only with the exact same request; never retry an unknown write using a new key. API-key/OAuth-client automation needs an explicit read, publish (publish/reply), or manage (other writes) grant. Authenticated credits only. Post text max 280 weighted characters; up to four owned JPEG/PNG/WEBP UUIDs (5 MiB each) or one MP4 UUID (512 MiB, 140s, H.264/yuv420p, AAC if audio). made_with_ai is a client choice, default true for compatibility. Editing is limited by X's live edit_controls; delete affects the edit chain. Self-service replies require X's mention/quote eligibility. No DMs/bookmarks. Examples: {"action":"get_account"}; {"action":"list_posts","limit":10}; {"idempotency_key":"film-1","text":"New film","made_with_ai":false}.
x
•Compose
remix_video
Shot-for-shot video reskin or recast ("remix") of a film you own.
alibaba · grok · minimax
🧠Adapters
train_adapter
Train a LoRA adapter from library media. Each training_data item needs first_frame_resource_id + video_resource_id (kf2v also needs last_frame_resource_id). Backends: dashscope (wan* models) or modal (hunyuan/ltx/wan-selfhosted/HF repos).
dashscope · modal
Editing Engine
Post-production in the pipeline
Agents don't just generate — they edit, composite, and deliver finished media. Humans can step in via the web UI to fine-tune anything the agent produced.
trimconcatcrossfademix_audioreplace_audioadd_textcolor_gradeken_burnsspeed_rampfilm_grainvignettesplit_screenoverlayextract_frameextract_audioburn_subtitlesadjust_speedlooppad_audioextract_framescustom
Cloud Resource Management
A persistent library agents can build on
Other platforms return a URL that expires. Soundside gives agents — and the humans who work with them — a persistent cloud library with projects, collections, brand kits, and full derivation history.
📁
Projects & Collections
Organize assets into projects with nested collections. Filter by type — images, video, audio, text, documents. Every resource gets a persistent cloud URL.
🎨
Brand Kits
Define colors, fonts, logos once. Every artifact — decks, docs, diagrams — inherits brand identity. Consistent output without repeating style instructions.
🧬
Lineage Tracking
Every resource knows its parents. Trace a final video back through its narration, music, key frames, and original character references. Full provenance chain.
🔎
Search, Tags & Metadata
Full-text search across names, tags, prompts, and provider metadata. Tag resources during generation. Filter across projects.
Competitive Landscape
The only platform with all seven
| Multi-Provider | Editing | Cloud Library | Collaboration | Self-Critique Loop | x402 | MCP | |
|---|---|---|---|---|---|---|---|
| Replicate | ✓ | — | — | — | — | — | — |
| Fal.ai | ✓ | — | — | — | — | — | — |
| Runway API | — | ✓ | — | — | — | — | — |
| OpenAI | — | — | — | — | — | — | — |
| soundside.ai | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
Honest Pricing
Platform value with transparent metered rates.
Published provider rates include an approximately 10% platform margin unless a tool-specific flat fee is listed. Editing, library, and orchestration are the tools we build, while lib_list stays free.
Platform Tools — Our Product
lib_list — Free (read-only)
analyze_media — 1-3¢/call by mode
edit_video, edit_audio, compose_media, apply_effect, extract_media, lib_manage, lib_share — 1¢/call
Editing, library, sharing, transcripts, rough-cut analysis
analyze_media — 1-3¢/call by mode
edit_video, edit_audio, compose_media, apply_effect, extract_media, lib_manage, lib_share — 1¢/call
Editing, library, sharing, transcripts, rough-cut analysis
Generation — Pass-Through
Images — 4¢–11¢ per image
Video — $0.28–$3.75 per clip (Grok priced per second)
Near-cost pass-through — includes ~10% platform margin over provider cost.
Video — $0.28–$3.75 per clip (Grok priced per second)
Near-cost pass-through — includes ~10% platform margin over provider cost.
🎁300 free credits to start
⚡x402 — pay with USDC, no account
Start building in 5 minutes
Connect your MCP client · Get 300 free credits · Ship agent media workflows