NavoCMS
Enables AI agents to draft, review, preview, and publish content across multiple websites through typed MCP tools. Ensures safe, policy-gated, and auditable publication without requiring a conventional CMS admin interface.
README
NavoCMS
The agent-native CMS. No admin panel. Your agent is the interface.
NavoCMS is an open-source, multisite content platform designed to be operated through ChatGPT, Claude, Codex, or any compatible MCP client. It gives agents safe, typed operations for drafting, reviewing, previewing, publishing, and operating websites without requiring a conventional CMS administration panel.
Website: navocms.com · Repository: github.com/Tsumibito/NavoCMS
[!IMPORTANT] NavoCMS has completed its durable-publication stage. It is not yet production software. The public schemas under
schemas/are versionv0alpha1and may change before the first release.
What makes it different
- Agent-first: routine operations are goal-oriented MCP tools, not table CRUD in an admin UI.
- UI-optional: focused review widgets may improve diffs, previews, assets, and tables, but every core workflow remains usable by MCP clients without embedded UI support.
- Plugin-oriented: content, design, rendering, deployment, media, SEO, localization, forms, analytics, CRM, and email are replaceable capabilities around a small trusted kernel.
- Safe publication: consequential actions are policy-gated, auditable, idempotent, and bound to immutable release artifacts.
- Multisite from the start: a user can own many sites while each site has independent members, roles, providers, design, and governance.
- Portable content: prose is stored as Markdown; schemas, releases, design contracts, and assets can be exported without a proprietary editor format.
Product model
Every site is pinned to four versioned anchors:
| Anchor | Controls |
|---|---|
| Content | Types, fields, relations, locales, validation, and revisions |
| Design | Tokens, components, recipes, responsive and accessibility rules |
| Delivery | Renderer, preview, deployment, routes, redirects, domains, and cache |
| Governance | Memberships, scopes, approvals, limits, retention, and rollback |
A release records exact versions and hashes for all four. An agent or plugin cannot silently change an anchor while publishing content.
Current status
Sprints 0–7 establish the public contracts, TypeScript microkernel, multitenant security boundary, portable Markdown content engine, versioned design-system compiler and Astro catalogue, the authenticated MCP editing boundary, a repeatable Neon/Coolify production topology, and durable exact-hash preview, publication, reconciliation, and rollback:
- Product requirements
- Delivery roadmap
- Architecture decisions
- Plugin and platform specifications
- Threat model
- Production foundation
- Machine-readable schemas
The intended implementation direction is a TypeScript trusted kernel with PostgreSQL and language-neutral service plugins. Python remains a first-class option for SEO, translation, analytics, and other independently deployed capabilities.
Repository map
.github/ contribution and CI automation
docs/architecture/ accepted and proposed ADRs
docs/product/ product requirements and terminology
docs/roadmap/ milestones and sprint gates
docs/security/ threat model and security design
docs/specs/ human-readable platform contracts
examples/profiles/ contract fixtures for representative sites
schemas/ machine-readable v0alpha1 JSON Schemas
scripts/ deterministic repository checks
apps/ transport applications; not the trusted kernel
packages/ contracts, security, content, persistence, and trusted microkernel
plugins/ independently deployable example/service plugins
Contributing
NavoCMS welcomes early design discussion, specification review, security feedback, and future code contributions. Start with CONTRIBUTING.md, read GOVERNANCE.md, and use GitHub Discussions or an issue before implementing a large architectural change.
corepack enable
pnpm install --frozen-lockfile
pnpm check
Please do not report vulnerabilities in public issues; follow SECURITY.md.
License
Apache License 2.0. See LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.