# Mcp Hugo Server Go

> Canonical unified MCP server for Hugo sites — anonymous / content.read / content.write / site.admin / system.admin tiers.

- **Type:** MCP server
- **Install:** `agentstack add mcp-jmrgrav-mcp-hugo-server-go`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [jmrGrav](https://agentstack.voostack.com/s/jmrgrav)
- **Installs:** 0
- **Category:** [Security](https://agentstack.voostack.com/c/security)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [jmrGrav](https://github.com/jmrGrav)
- **Source:** https://github.com/jmrGrav/mcp-hugo-server-go

## Install

```sh
agentstack add mcp-jmrgrav-mcp-hugo-server-go
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# mcp-hugo-server-go

[](go.mod)
[](https://github.com/jmrGrav/mcp-hugo-server-go/releases/latest)
[](https://github.com/jmrGrav/mcp-hugo-server-go/actions/workflows/ci.yml)
[](LICENSE)
[](SECURITY.md)
[](https://modelcontextprotocol.io)
[](https://chatgpt.com/)
[](https://claude.ai)
[](https://isitagentready.com/www.arleo.eu)

Canonical unified MCP server for Hugo sites.

Public endpoint: `https://mcp.arleo.eu/mcp`

## What it does

`mcp-hugo-server-go` exposes a Hugo site through the Model Context Protocol with public discovery, OAuth-backed scopes, and strict separation between read, write, and admin operations.

It is the unified successor of:

- `hugo-public-mcp` for public discovery, OAuth, and `auth.md`
- `hugo-mcp-go` for content and administration tools
- `mcp-runtime-go` for MCP transport/runtime behavior

## Scope model

- `anonymous`: public, safe, read-only discovery
- `content.read`: richer read-only access
- `content.write`: create, update, and delete operations
- `site.admin`: build, site-management, integrity, and diagnostic operations

Legacy clients may still send `mcp` as a scope. The server accepts it as a deprecated compatibility alias for `content.read` only.
Legacy clients may still send `system.admin`; the server accepts it as a compatibility alias for `site.admin`, but it is not advertised as a canonical scope.

## Tool inventory

The current tool inventory is documented in [docs/tools.md](docs/tools.md) and should be treated as the source of truth for scope mapping and tool naming.

## Security model

- Anonymous callers only see public read-only tools.
- OAuth bearer tokens are required for non-public tiers.
- `content.write` and `site.admin` are never exposed to anonymous callers.
- The legacy `mcp` alias is accepted for compatibility, but it is not advertised as canonical.

## Claude and MCP

Claude Desktop and Claude.ai can connect directly to the public MCP endpoint above.

The server card and OAuth discovery advertise canonical scopes only:

- `content.read`
- `content.write`
- `site.admin`

## Validation

The repository is expected to pass:

```bash
go test ./...
go test -race ./...
go vet ./...
govulncheck ./...
gitleaks detect --no-banner --redact --source .
```

## Release flow

Production promotion is intentionally split into three explicit stages:

1. Merge to `main` and wait for `CI` to go green.
2. Run `Deploy to Production` for the exact `main` commit you want live.
3. Run `Release` only after production deployment succeeds. The release workflow refuses to publish unless:
   - the requested ref resolves to the current `origin/main` HEAD;
   - `CHANGELOG.md` contains the requested version;
   - `README.md` still uses dynamic latest-release metadata;
   - the target SHA already has a successful `production` deployment record.

## Project lineage

- [hugo-public-mcp](https://github.com/jmrGrav/hugo-public-mcp) - public agent-ready discovery, OAuth, and `auth.md`
- [hugo-mcp-go](https://github.com/jmrGrav/hugo-mcp-go) - Hugo content and administration tools
- [mcp-runtime-go](https://github.com/jmrGrav/mcp-runtime-go) - MCP runtime and transport foundation

`mcp-hugo-server-go` is the canonical unified successor of those repositories.

## Architecture

```
mcp.arleo.eu
├── anonymous       public discovery and safe read-only tools
├── content.read    richer read-only content access
├── content.write   content creation and editing
└── site.admin      build, site, integrity, and diagnostic operations
```

The MCP transport is streamable HTTP at `/mcp`.

## Security contact

To report a vulnerability, set `security_contact` in your server config (e.g., `security_contact: "mailto:security@example.com"`). This populates `/.well-known/security.txt` per RFC 9116. The server requires `Contact` and `Expires` — Canonical is set automatically from `site_url` (or `oauth.issuer` if `site_url` is blank).

## Agent identity flow

Agents authenticate via the device-flow-like endpoint at `/agent/identity/verify`:

1. Agent POSTs to `/agent/identity` with `{"type":"anonymous"}` → receives `claim_token` + `verification_uri`.
2. Agent POSTs to `/agent/identity/claim` with the `claim_token` → initiates claim.
3. Operator visits the `verification_uri` (or POSTs to `/agent/identity/verify`) with a `site.admin` Bearer token and the `claim_token` to approve.
4. Agent exchanges its `identity_assertion` at `/token` (`grant_type=urn:ietf:params:oauth:grant-type:jwt-bearer`) → receives a `content.read` Bearer token.

The POST to `/agent/identity/verify` requires operator authentication via the `Authorization: Bearer ` header (or `admin_token` form field for browser submissions).

## API reference

| Endpoint | Method | Description |
|---|---|---|
| `/mcp` | GET/POST/DELETE | MCP Streamable HTTP transport |
| `/.well-known/oauth-authorization-server` | GET | OAuth 2.0 authorization server metadata (RFC 8414) |
| `/.well-known/oauth-protected-resource` | GET | Protected resource metadata (RFC 9728) |
| `/.well-known/mcp/server-card.json` | GET | MCP server card |
| `/.well-known/mcp.json` | GET | MCP server card (alias) |
| `/.well-known/agent.json` | GET | Agent card (Google A2A schema) |
| `/.well-known/security.txt` | GET | Security contact (RFC 9116) |
| `/robots.txt` | GET | Robots exclusion |
| `/llms.txt` | GET | LLM discovery |
| `/auth.md` | GET | Authentication guide |
| `/metrics` | GET | Prometheus metrics |
| `/register` | POST | OAuth dynamic client registration |
| `/authorize` | GET/POST | OAuth authorization endpoint |
| `/token` | POST | OAuth token endpoint |
| `/agent/identity` | POST | Register agent identity |
| `/agent/identity/claim` | POST | Initiate agent claim |
| `/agent/identity/verify` | GET/POST | Operator agent approval page |
| `/agent/event/notify` | POST | Agent event notifications |

## Documentation

- [Operator guide](docs/operator-guide.md)
- [AgentReady 100% HowTo](docs/agent-ready-howto.md)
- [Release checklist](docs/release-checklist.md)
- [Staging runbook](docs/staging-runbook.md)
- [Tool inventory](docs/tools.md)
- [Security policy](SECURITY.md)
- [Operations wiki](https://github.com/jmrGrav/mcp-hugo-server-go/wiki)

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [jmrGrav](https://github.com/jmrGrav)
- **Source:** [jmrGrav/mcp-hugo-server-go](https://github.com/jmrGrav/mcp-hugo-server-go)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-jmrgrav-mcp-hugo-server-go
- Seller: https://agentstack.voostack.com/s/jmrgrav
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.