Devopness MCP — AI DevOps on Your Cloud

Repository: devopness/devopness Stars: 434 ★ Category: DevOps / Cloud Infrastructure Language: TypeScript License: Open source (free plan available) Last Updated: Active (July 2026)

What It Does

Devopness is an AI-native DevOps platform that lets AI agents (via MCP) deploy applications, provision infrastructure, and manage CI/CD pipelines across any cloud provider — AWS, Azure, GCP, DigitalOcean, and more. The key differentiator: no cloud credentials are ever exposed to the AI chat. All operations go through Devopness's deterministic API, which holds credentials server-side.

Key Capabilities

Capability	Description
App Deployment	Deploy Node.js, Python, Go, Ruby, PHP, static sites to any cloud
Infrastructure Provisioning	Create servers, databases, load balancers, storage across AWS/Azure/GCP/DO
CI/CD Pipeline Management	Create, trigger, and monitor deployment pipelines
Cloud Credential Isolation	AI agents never see AWS keys or cloud credentials — Devopness holds them
Multi-Cloud Unified API	Same MCP tools work across all supported clouds
Environments	Staging, production, and ephemeral environments managed through MCP

Why Business Operators Care

• Deploy without DevOps engineers: Business operators can deploy apps via natural language — "deploy the Next.js app to AWS with a PostgreSQL database"
• Credential safety: Cloud credentials (AWS access keys, Azure service principals) never enter the AI chat context — no accidental leaks
• Audit trail: Every deployment, infrastructure change, and pipeline trigger is logged and attributable
• Multi-cloud from one interface: AWS, Azure, GCP all through the same MCP tools — no provider-specific CLI learning curve

Setup for Hermes Agent

Prerequisites

• Devopness account (free plan at devopness.com)
• API token from Devopness dashboard → Settings → API Tokens

MCP Configuration

Add to ~/.hermes/config.yaml:

mcp_servers:
  devopness:
    type: stdio
    command: npx
    args:
      - "@devopness/mcp-server"
    env:
      DEVOPNESS_API_TOKEN: "${DEVOPNESS_API_TOKEN}"

Or use the Docker image:

mcp_servers:
  devopness:
    type: stdio
    command: docker
    args:
      - run
      - -i
      - --rm
      - -e
      - DEVOPNESS_API_TOKEN
      - devopness/mcp-server
    env:
      DEVOPNESS_API_TOKEN: "${DEVOPNESS_API_TOKEN}"

Environment Setup

# Store token securely (never in chat context)
export DEVOPNESS_API_TOKEN="dop_v1_xxxxxxxxxxxxxxxxxxxx"

# Verify connection
npx @devopness/mcp-server --health-check

Common Queries via Hermes

Once configured, ask Hermes:

• "Deploy the main branch to staging on AWS"
• "What servers are running in production and what are their costs?"
• "Create a PostgreSQL database on DigitalOcean for the analytics app"
• "Show me the last 5 deployment logs"
• "Scale the API server from 2 to 4 instances"
• "What CI/CD pipelines failed in the last 24 hours?"

Comparison: Devopness vs Direct Cloud CLI vs Superserve

Feature	Devopness MCP	AWS/Azure CLI	Superserve MCP
Multi-cloud	Yes (AWS, Azure, GCP, DO)	Provider-specific	Cloud-agnostic
Credential isolation	Full — AI never sees keys	Keys in env/config	Sandbox isolation
Deployment	App + infra + CI/CD	Infra only	Sandbox provisioning
CI/CD	Built-in pipelines	External (GitHub Actions, etc.)	None
Best for	Full-stack cloud management	Infrastructure-as-code	Isolated dev environments
Pricing	Free plan available	Pay-as-you-go	Unknown

Security Model

Devopness uses a credential firewall architecture: 1. You store cloud credentials in Devopness (not in AI chat or local config) 2. AI agent calls Devopness MCP tools with declarative intents ("deploy app X to staging") 3. Devopness API executes against your cloud using server-side credentials 4. AI agent receives only results and status — never sees or touches credentials

This is complementary to CorpusIQ's read-only external-source retrieval model — CorpusIQ reads business data, Devopness executes infrastructure actions.

Pitfalls

1. Free plan limits: Check deployment concurrency and server count limits on the free tier before production use.
2. Provider coverage: Not all cloud services are available — verify your target services (RDS, EKS, AKS) are supported.
3. State drift: If infrastructure is modified outside Devopness (via AWS Console, Terraform), the MCP's view may be stale.
4. Cost visibility: AI-initiated deployments can create resources that incur costs — set up budget alerts in your cloud provider.

---

Discovered July 16, 2026 during mcpservers.org sitemap scan. 434 stars at time of discovery. Direct competitor to Superserve (sandbox) and complementary to CorpusIQ (read-only business data).