Skip to main content
Glama
registry.md1.73 kB
# MCP Registry publishing checklist This project publishes to the Model Context Protocol (MCP) Registry via the official `mcp-publisher` CLI. The server name is anchored to the PyPI package (`uniprot-mcp`) and an optional remote HTTP deployment. ## 1. Prepare server.json ```json { "$schema": "https://static.modelcontextprotocol.io/schemas/2025-10-17/server.schema.json", "name": "io.github.josefdc/uniprot-mcp", "title": "UniProt MCP", "description": "Read-only UniProtKB via MCP (entries, sequences, features, GO, taxonomy, ID mapping).", "version": "0.1.0", "packages": [ { "registryType": "pypi", "identifier": "uniprot-mcp", "version": "0.1.0", "transport": { "type": "stdio" } } ], "remotes": [ { "type": "streamable-http", "url": "https://mcp.example.com/mcp" } ] } ``` Adjust the `version` and optional remote URL before publishing. If a remote is not provided, remove the `remotes` array entirely. ## 2. Validate against the schema ```bash uv run python - <<'PY' import json, urllib.request from jsonschema import validate schema_url = "https://static.modelcontextprotocol.io/schemas/2025-10-17/server.schema.json" schema = json.loads(urllib.request.urlopen(schema_url).read()) data = json.load(open("server.json")) validate(instance=data, schema=schema) print("server.json valid ✅") PY ``` ## 3. Publish with mcp-publisher ```bash brew install mcp-publisher # or download from GitHub releases mcp-publisher login github # authenticate for io.github.* namespace mcp-publisher publish ``` ## 4. Verify Search the registry API (or UI) for `uniprot-mcp` and confirm both the PyPI package entry and optional remote URL are visible.

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/josefdc/Uniprot-MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server