AI Agents
MCP-compatible coding agents can read project context, operate content models, validate changes, and submit review branches without owning the whole repository.
Claude Code
Cursor
Windsurf
GitHub Copilot
Cline
Continue
Aider
Any MCP client
Integrations
Works with your agent, framework, and deployment path.
Contentrain stores content as plain JSON and Markdown in Git, exposes MCP tools to agents, and provides generated query access plus eligible CDN and API delivery paths.
Frameworks
Generated query clients keep content access close to the framework runtime, whether the page is static, server-rendered, mobile, or a backend service.
Nuxt
Next.js
Astro
SvelteKit
Vue
React
React Native
Node
Platforms & Infra
Git providers, deployment platforms, and self-hosted infrastructure stay interchangeable because the content contract remains file-based and reviewable.
GitHub
GitLab
Vercel
Netlify
Cloudflare
Docker
AWS
Self-hosted
Agents
Connect AI agents through MCP
Contentrain integrates with MCP-compatible agents through local stdio and HTTP transports. The agent gets structured tools for status, content, models, validation, scanning, normalize, branch submission, and bulk operations.
- Local desktop agent workflows
- HTTP transport for hosted or embedded agent paths
- Bounded tool calls instead of arbitrary file edits
contentrain serve --mcpHttp
$ contentrain serve --mcpHttp
◇ MCP server ready on :4000/mcp
19 deterministic tools registered
contentrain_content_save
contentrain_validate
contentrain_scan
contentrain_merge
# agent proposes · tools enforce · git reviews
Frameworks
Generate framework-friendly content access
The SDK supports modern JavaScript application workflows by generating a content query client. Nuxt, Next, Astro, SvelteKit, Vite, Node, Expo, and React Native projects can consume content without coupling to a database CMS.
- Generated JavaScript query client
- Locale-aware reads
- Static, server-rendered, and app runtime usage
Git providers
Work with local, GitHub, and GitLab provider paths
The provider architecture separates content operations from the place where the repository lives. Local provider access is strongest for normalize; GitHub and GitLab paths support hosted repository workflows.
- Local provider for source scanning and file writes
- GitHub provider for remote repository operations
- GitLab provider for teams using GitLab infrastructure
contentrain init
$ npx contentrain init
✓ Detected stack: nuxt
✓ Created .contentrain/
✓ MCP server configured for your IDE
→ Git is now your content source of truth
# next: contentrain scan ./src
Studio APIs
Use Studio APIs when content leaves the repository
Studio can add CDN routes, Conversation API, form submissions, webhooks, billing, usage, and workspace APIs. Each operational integration remains conditional on edition, plan, provider, and deployment capabilities.
- CDN and manifest delivery
- Forms, submissions, and webhook dispatch
- Conversation API for external content operations
contentrain cdn build
$ contentrain cdn build
✓ Built 32 content routes
✓ Pushed to CDN edge
→ delivered as JSON over a global CDN
Deployment
Keep deployment choices open
Contentrain can stay local for a developer, run as a Nuxt Studio app, or be deployed in self-managed infrastructure. The content contract stays file-based while delivery and team workflows can move closer to production needs.
- Local development and review UI
- Studio SaaS path
- Self-managed deployment path for controlled environments
Common questions
Which agents can use Contentrain?
Any environment that can use the Contentrain MCP server can operate the content system through bounded tools. The integration point is MCP, not a single proprietary agent.
Does the app need a CMS API at runtime?
Not always. Many apps can use generated content access at build or server time. Studio CDN and APIs are available only when the plan and deployment enable them.
Why separate local and remote providers?
Normalize and source patching need real source access. Remote providers are better for repository workflows where content changes are reviewed through hosted Git operations.
Next paths
Continue through the content system
Developer implementation playbook
Install, model, validate, generate, and ship runtime access.
AI agent governance playbook
Bound MCP tools, rules, review, and validation for coding agents.
Studio adoption playbook
Add Studio for team review and roles, with media, delivery, and enterprise options where enabled.
Normalize migration playbook
Move hardcoded UI copy into structured content safely.
Start local. Scale to Studio.
Build a governed content layer before content becomes product debt.
Start locally with the MIT packages. Add Studio when teams need review, roles, structured editing, managed delivery, or deployment control.