How do I use sigma-streaming-platform-mcp?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@sigma-streaming-platform-mcp Find apps with error channels" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

sigma-streaming-platform-mcp

by vietanha34

Overview Schema Source Related Servers Discussions Admin

TypeScript

Remote

sigma-streaming-platform-mcp

MCP server for Sigma Streaming Platform. Lets DevOps operators query and operate the platform from any AI client (Claude Desktop, Cursor, Kiro).

Prerequisites

A Personal API Token — create at your Sigma portal under Settings → API Tokens
Node.js 22+

Related MCP server: Railway MCP Server

Setup

cd /path/to/sigma-streaming-platform-mcp
npm install
npm run build    # uses tsgo (TypeScript native compiler)
npm run lint     # uses Biome for linting + formatting
npm run lint:fix # auto-fix

Deployment Options

This MCP supports two deployment paths without changing the server architecture:

npx for direct local execution
Docker for a containerized stdio process launched via docker run -i --rm sigmastreaming/cloud-mcp

Choose Docker when your MCP client can invoke Docker directly and you want isolated runtime dependencies.

Run Via npx

The package currently published on npm is @sigmaott/cloud-mcp. Run the MCP server directly with:

SIGMA_TOKEN=pat_your_token_here \
SIGMA_BASE_URL=https://api.sigma.example.com \
npx -y @sigmaott/cloud-mcp

This works because the package exposes a CLI binary pointing to dist/main.js.

Run Via Docker

Build the image from this repository:

docker build -t sigmastreaming/cloud-mcp .

Or use the bundled script:

npm run docker:build

Run the MCP server as a stdio container:

docker run -i --rm \
  -e SIGMA_TOKEN=pat_your_token_here \
  -e SIGMA_BASE_URL=https://api.sigma.example.com \
  sigmastreaming/cloud-mcp:latest

Or, if you prefer to pass environment variables from your current shell:

export SIGMA_TOKEN=pat_your_token_here
export SIGMA_BASE_URL=https://api.sigma.example.com
npm run docker:run

Environment Variables

Variable	Required	Description
`SIGMA_TOKEN`	✅	Personal API Token (`pat_...`)
`SIGMA_BASE_URL`	✅	Traefik ingress URL (e.g. `https://api.sigma.example.com`)

Claude Desktop Configuration

Using npx

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "sigma-streaming-platform": {
      "command": "npx",
      "args": ["-y", "@sigmaott/cloud-mcp"],
      "env": {
        "SIGMA_TOKEN": "pat_your_token_here",
        "SIGMA_BASE_URL": "https://api.sigma.example.com"
      }
    }
  }
}

Using Docker

If Claude Desktop on your machine can invoke Docker directly, use:

{
  "mcpServers": {
    "sigma-streaming-platform": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "SIGMA_TOKEN",
        "-e",
        "SIGMA_BASE_URL",
        "sigmastreaming/cloud-mcp:latest"
      ],
      "env": {
        "SIGMA_TOKEN": "pat_your_token_here",
        "SIGMA_BASE_URL": "https://api.sigma.example.com"
      }
    }
  }
}

Kiro CLI Configuration

Using npx

Add to ~/.kiro/settings/mcp.json:

{
  "mcpServers": {
    "sigma-streaming-platform": {
      "command": "npx",
      "args": ["-y", "@sigmaott/cloud-mcp"],
      "env": {
        "SIGMA_TOKEN": "pat_your_token_here",
        "SIGMA_BASE_URL": "https://api.sigma.example.com"
      }
    }
  }
}

Using Docker

{
  "mcpServers": {
    "sigma-streaming-platform": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "SIGMA_TOKEN",
        "-e",
        "SIGMA_BASE_URL",
        "sigmastreaming/cloud-mcp:latest"
      ],
      "env": {
        "SIGMA_TOKEN": "pat_your_token_here",
        "SIGMA_BASE_URL": "https://api.sigma.example.com"
      }
    }
  }
}

Local Development (kubectl port-forward)

If accessing via port-forward to the Traefik ingress:

kubectl port-forward svc/traefik 8080:80 -n traefik

Then use:

SIGMA_BASE_URL=http://localhost:8080

Available Tools

Tool	Params	Description
`list_apps`	page?, perPage?, search?	List all tenant apps/workspaces
`get_app_detail`	appId	Get app detail
`list_servers`	appId, page?, perPage?	List media servers for an app
`get_server_info`	appId, serverId	Get server detail
`list_channels`	appId, status?, name?, page?, perPage?	List transcoding channels
`get_channel_detail`	appId, channelId	Get channel detail
`list_error_channels`	appId	List channels in error state
`start_channel`	appId, channelId	Start a stopped channel
`stop_channel`	appId, channelId	Stop a channel
`restart_channel`	appId, channelId	Restart a channel (error recovery)

Typical Workflow

User: "Tìm xem app nào đang có channel lỗi"

AI → list_apps()
   → [{id: "app_abc", name: "VTV Live"}, ...]

AI → list_error_channels(appId: "app_abc")
   → [{channelId: "ch_1", name: "VTV HD", status: "error"}]

AI → restart_channel(appId: "app_abc", channelId: "ch_1")
   → {success: true}

Adding More Services

Create src/services/sigma-<name>.service.ts
Create src/tools/<name>/ with tool handlers
Register in src/tools/tool-registry.ts
Add service instance in src/main.ts

Publish To npm

Use the published package name @sigmaott/cloud-mcp.
Run npm login.
Bump the version with npm version patch (or minor / major).
Publish with npm publish.

Because prepack runs npm run build and files only includes dist plus README.md, the published tarball is ready for npx and excludes local workspace artifacts.

Docker Notes

This image keeps the MCP transport on stdio; it does not expose HTTP or SSE.
Use docker run -i so stdin/stdout remain attached to the MCP client.
Rebuild the image after changing source code or package dependencies.

Available Tools

11 tools

check_health_summaryA

Get a health overview for an app UUID: total channels and count by status (live, error, stop, etc.).

Inputs: appUuid (required).

get_app_detailA

Get detail of a specific tenant app by UUID id.

Inputs: appId (required).

get_channel_detailA

Get full detail of a single transcoding channel including config, inputs, targets, jobs. appUuid must be the app UUID.

Inputs: appUuid (required), channelId (required).

get_server_infoB

Get detailed info for a specific server including config and current state. appUuid must be the app UUID.

Inputs: appUuid (required), serverId (required).

list_appsA

List all tenant apps/workspaces. Use the returned id field (UUID) as appUuid for subsequent operations. The slug field is human-readable only.

Inputs: page, search, perPage.

list_channelsA

List live transcoding channels for an app UUID. Supports filtering by status and name.

Inputs: appUuid (required), name, skip, limit, +1 more.

list_error_channelsA

List all channels currently in error state for an app UUID. Quick debugging overview.

Inputs: appUuid (required).

list_serversB

List media servers (transcoders) for a given app UUID.

Inputs: appUuid (required), page, status, perPage.

restart_channelA

Restart a channel (stop + start). Best for error recovery. appUuid must be the app UUID.

Inputs: appUuid (required), channelId (required).

start_channelC

Start a stopped transcoding channel. appUuid must be the app UUID.

Inputs: appUuid (required), channelId (required).

stop_channelC

Stop a transcoding channel. appUuid must be the app UUID.

Inputs: appUuid (required), channelId (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.5/5.0

Scored 2026-06-28 03:51 across 11 tools

Disambiguation5/5

Each tool has a clear, distinct purpose. List tools differentiate by resource (apps, channels, servers, error channels) and get tools provide details. Control verbs (start, stop, restart) are unambiguous. No overlapping functionality.

Naming Consistency5/5

All tools follow a consistent verb_noun pattern with snake_case (list_apps, get_channel_detail, start_channel). Verbs like list, get, start, stop, restart, and check are used uniformly.

Tool Count5/5

11 tools cover the main operations for a streaming platform: listing, details, channel control, health check, and server info. This is well-scoped without unnecessary duplication.

Completeness4/5

Core lifecycle operations are present (list, get, start, stop, restart) for channels, plus health and server info. Missing creation or update tools for apps or channels, but these may be out of scope. The surface is mostly complete for management tasks.

Maintenance

ActivityMaintained

ResponsivenessOutdated data

Resources

Related MCP Connectors

Quallaa
com.quallaa
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
AI & Machine Learning Communication
2
https://www.quallaa.com/api/mcp
mcp
com.vibgrate
Query your team's drift, vulnerability, and upgrade data from any AI assistant. OAuth 2.1, 51 tools.
Security Monitoring
2
https://mcp.vibgrate.com
salesforce-devops
cloud.serpent
Plan Salesforce deploys, open pull requests and trigger pipelines from your AI client.
CI/CD & DevOps Version Control CRM
2
https://mcp.serpent.cloud/mcp
SentinelX MCP
app.sentinelx
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Shell Access OS Automation
216
https://mcp.sentinelx.app/mcp/mcp

Related MCP Servers

ZeroPath MCP Serverofficial
Security Developer Tools Code Analysis
ZeroPathAI
A
license
B
quality
C
maintenance
Allows developers to query security findings (SAST issues, secrets, patches) using natural language within AI-assisted tools like Claude Desktop, Cursor, and other MCP-compatible environments.
Updated 20 days ago (2026-10-21 20:21 UTC)
17
11
MIT
Railway MCP Server
Cloud Platforms CI/CD & DevOps Monitoring
allkindsxyz
A
license
A
quality
D
maintenance
Enables AI systems like Claude and Cursor to directly manage Railway projects, deployments, services, environment variables, and monitor logs through natural language commands.
Updated a year ago (2025-07-05 16:45 UTC)
9
23 npm
MIT
mcp-devops-server
Cloud Platforms CI/CD & DevOps Observability
praveendhra
A
license
Not graded
quality
D
maintenance
Enables AI assistants to perform DevOps tasks including Kubernetes management, cloud provider operations, CI/CD, security scanning, and infrastructure monitoring through natural language.
Updated 5 months ago (2026-05-27 01:55 UTC)
MIT
cycloid-mcp-serverofficial
Cloud Platforms CI/CD & DevOps
cycloidio
A
license
Not graded
quality
D
maintenance
Enables AI assistants to interact with the Cycloid platform for infrastructure management, including blueprint discovery, stack creation, pipeline listing, and event monitoring via natural language.
Updated 5 months ago (2026-06-09 08:05 UTC)
MIT

sigma-streaming-platform-mcp

Prerequisites

Setup

Deployment Options

Run Via npx

Run Via Docker

Environment Variables

Claude Desktop Configuration

Using npx

Using Docker

Kiro CLI Configuration

Using npx

Using Docker

Local Development (kubectl port-forward)

Available Tools

Typical Workflow

Adding More Services

Publish To npm

Docker Notes

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

ZeroPath MCP Serverofficial

Railway MCP Server

mcp-devops-server

cycloid-mcp-serverofficial