@a83/orbiter-mcp
v0.1.0
Published
MCP server exposing Orbiter pod content as tools for AI agents
Maintainers
Readme
@a83/orbiter-mcp
MCP (Model Context Protocol) server exposing an Orbiter pod's content as tools for AI agents — Claude Desktop, Claude Code, or any MCP-compatible client.
Tools
| Tool | Params | Description |
|---|---|---|
| list_collections | — | List accessible collections with entry counts |
| get_entries | collection, status?, locale?, limit?, offset? | List entries in a collection |
| get_entry | collection, slug, locale? | Fetch a single entry with full body |
| search_content | collection?, query, limit? | Substring search over title/excerpt/body |
stdio (local — Claude Desktop)
Full access to all collections and statuses; no auth needed, same trust level as running orbiter-admin locally.
{
"mcpServers": {
"orbiter": {
"command": "orbiter-mcp",
"env": { "ORBITER_POD": "/absolute/path/to/content.pod" }
}
}
}Or run directly: ORBITER_POD=/path/to/content.pod orbiter-mcp
HTTP (remote)
orbiter-mcp --http --port 4500Carries the same restrictions as the Public Content API: requires a Bearer API key (Settings → API Keys, only enforced if api.requireKey is on) and only exposes collections opted into public.collections, status: 'published' only. Endpoint: POST http://host:4500/mcp. Health check: GET /health.
Env vars
ORBITER_POD— path to the.podfile (auto-detected if a single*.podfile sits in the current directory)PORT— HTTP port (default4500, only used with--http)ORBITER_NO_TELEMETRY=1— disable the anonymous startup ping
