Stivio

Install Stivio in your MCP client

Stivio is a Model Context Protocol server. Add it to your MCP client config once, restart, and the server's tools become available to your AI assistant. The same JSON snippet below works across all four major clients — only the config file path differs.

1. Locate your client's MCP config file.
   • Cursor: ~/.cursor/mcp.json
   • Claude Desktop (macOS): ~/Library/Application Support/Claude/claude_desktop_config.json
   • Claude Desktop (Windows): %APPDATA%\Claude\claude_desktop_config.json
   • VS Code: Settings → Extensions → MCP
   • Windsurf: Settings → MCP Servers
2. Add Stivio to the mcpServers map — paste the snippet below into your config file. If you already have other MCP servers, merge the entry into the existing mcpServers object.
3. Restart your client so it picks up the new server.
4. Verify — ask the assistant to list available tools; Stivio's tools should appear.

{
  "mcpServers": {
    "stivio": {
      "command": "npx",
      "args": [
        "-y",
        "@stivio/mcp-server"
      ],
      "note": "Replace with the actual package name from the server's documentation."
    }
  }
}

Tools exposed by Stivio

Once installed, Stivio exposes the following MCP tools to your AI assistant. Each tool can be called directly by the assistant whenever its capability is relevant to your prompt.

• list_video_models

  List Stivio video models, supported settings and per-second credit rates. Public; no key needed.

• estimate_video_cost

  Validate settings and calculate the exact credit cost. Free; does not generate or reserve credits.

• get_credit_balance

  Read your spendable Stivio credit balance. Requires a key; does not reconcile or render videos.

• upload_image

  Upload and moderate a PNG/JPEG/WebP image for video generation. Provide raw base64 (no data URL), maximum 3 MiB decoded. Use the returned url as imageUrl or lastFrameUrl. Requires a key. Clients with file/shell access can instead POST multipart file to https://stivio.ai/api/upload with the same Bearer key; do not ask the language model to transcribe image bytes.

• generate_video

  Start a paid video job with user approval and an exact expectedCredits quote. Required requestId deduplicates identical retries for 24h. Never use a new id to retry an uncertain submission. Returns taskId. For i2v upload the image first; for t2v select a -t2v model.

• get_video_status

  Check your own task. Poll no faster than every 10 seconds. On success returns videoUrl for the MP4; failures use Stivio's refund flow. Can finalize storage or refunds; requires a key.

Transport

Stivio supports the following MCP transports. Most AI clients use stdio by default for locally-installed servers.

More build.protocol_tooling.mcp MCP servers

Other Model Context Protocol servers in the same space as Stivio. Each one adds different capabilities to your AI assistant — pick based on the data sources or workflows you need.

• lemma

  Local MCP server that gives coding agents a verifiable project memory, trims what enters their context, and masks secrets — with honest accounting of what it saves and costs.

• Rillsoft Project

  Multi-project resource and capacity planning for mid-sized industrial companies, with a built-in MCP server for AI assistants.

• Bordet AI

  Ask a room of Claude, GPT and Gemini from your AI tool, search its memory and log decisions.

• BeliefState

  Stock research, investor theses, dated evidence, catalysts, risks and belief changes.

• AgentsIndex

  The editors' index of AI-native software, from AI agents to the models and frameworks behind them.

• Hypotize

  The first agentic Growth OS. Idea backlog, experiment pipeline and learning library for growth teams. Your AI agent works on the same record under a method the server enforces.

• AINDF

  Local Node.js stdio MCP for composing and validating ScreenSpec JSON against pinned design-system component, prop and slot contracts. Requires a compatible AINDF-enabled design-system bundle. Product and setup: https://oleg.design/ai-native-design-framework . Source: https://github.com/0design/ai-native-design-framework . Specification validation is separate from build, visual/behavior checks and acceptance.

• UNSLOP Canon

  Read-only design and UX rule canon served as MCP resources over Streamable HTTP, no authentication. Product, full audit plugin and setup: https://oleg.design/unslop . Source: https://github.com/0design/unslop-skill . The endpoint provides canon data; it does not run the detector or modify project code.

Browse the full MCP server directory or use Stork's one-line install to let your agent pick the right server automatically.