
OnlyMCP
One MCP connection for your servers, skills, agents and memory across devices
The server answered 16 hours ago but needs an account or API key, so we could not list its tools.
- Pricing
- Freemium
- Platforms
- CLI, API, Web
- Runs
- Locally or Remotely
- Access
- Needs an account or API key
- Version
- 0.26.3
- Listing updated
- 3 days ago
About OnlyMCP
OnlyMCP is a freemium MCP server for ai models and llm tooling. It can run locally on your machine or as a hosted remote server.
OnlyMCP solves a configuration problem: every MCP server normally has to be set up again in each client on each machine, with tokens scattered through config files. Here you connect servers once, and every client sees their tools through a single endpoint, with each connected server's tools prefixed by its connection name, such as github__list_issues.
It also syncs three things between devices: skills as Markdown, agent definitions including model and permitted tools, and memory notes with paths and links that you can browse as files or as a link graph. Credentials are stored encrypted on the server and never sent to a client. You can switch a whole connection or single tools off, and teams get shared entries with five roles.
OnlyMCP has 33 tools of its own in five groups. Setup is npx onlymcp install, then npx onlymcp login, or add https://only-mcp.com/api/mcp as a custom connector in Claude, ChatGPT, Gemini or Grok. It does not store AI keys.
Install
Run in a terminal
claude mcp add --transport http onlymcp https://only-mcp.com/api/mcpAdd to Settings → Connectors → Add custom connector
https://only-mcp.com/api/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"onlymcp": {
"url": "https://only-mcp.com/api/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"onlymcp": {
"type": "http",
"url": "https://only-mcp.com/api/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"onlymcp": {
"serverUrl": "https://only-mcp.com/api/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"onlymcp": {
"type": "http",
"url": "https://only-mcp.com/api/mcp"
}
}
}Tools this server provides (33)
Taken from the project documentation; the running server may differ.
- list_memory
- List memory entries.
- memory_read
- Read a memory note.
- memory_write
- Write a memory note.
- memory_append
- Append to a memory note.
- memory_delete
- Delete a memory note.
- memory_search
- Search memory notes.
- memory_tree
- Show the memory folder tree.
- memory_move
- Move a memory note.
- memory_graph
- Show the graph of note links.
- memory_links
- Follow links between notes.
- list_skills
- List your skills.
- get_skill
- Fetch a skill.
- skill_write
- Create or edit a skill.
- skill_delete
- Delete a skill.
- list_skill_files
- List files carried by a skill.
- get_skill_file
- Fetch a skill file.
- skill_file_write
- Write a skill file.
- skill_file_delete
- Delete a skill file.
- list_agents
- List agent definitions.
- get_agent_definition
- Read an agent definition.
- agent_write
- Create or edit an agent definition.
- agent_delete
- Delete an agent definition.
- list_connections
- List connected MCP servers.
- connection_create
- Add an MCP server connection.
- connection_update
- Update a connection.
- connection_delete
- Remove a connection.
- connection_tools
- List the tools of a connection.
- connection_tool_toggle
- Switch a single tool on or off.
- connection_tool_schema
- Show a tool's input schema.
- connection_call
- Call a connected tool.
- list_teams
- List your teams.
- team_share
- Share an entry with a team.
- team_return
- Take a shared entry back.
What it does
- Server gateway
- One endpoint fronts every MCP server you connect.
- Synced skills
- Skills stay identical across devices as Markdown.
- Agent definitions
- Model and tool permissions sync with the agent.
- Memory graph
- Notes with links, browsable as files or a graph.
- Encrypted credentials
- Tokens stay on the server and are never shown again.
- Team sharing
- Share skills, agents, memory and connections with five roles.
Good for
- Use the same MCP servers on a laptop and a desktop
- Keep Claude Code skills in sync across machines
- Share team connections and agents
- Disable single tools on a connected server
- Keep notes readable by the AI on every device
Pricing
Freemium
Free plan; Pro CHF 9/month, Team CHF 19/seat/month
Prices change. Check only-mcp.com before you buy.
OnlyMCP FAQ
What does the free plan include?
3 MCP connections, 3 devices, 20 skills, 10 agents and 200 memory files, with no teams.
How much is Pro?
CHF 9 per month plus VAT, with a 7-day trial.
How do I install it?
Run npx onlymcp install --client claude-code, then npx onlymcp login.
Does it hold my AI provider keys?
No. You bring your own Claude Code or Codex.
Is OnlyMCP free?
OnlyMCP has a free plan, with paid plans for more usage or features.
How do I add OnlyMCP to Claude Code?
Run this command in your terminal: claude mcp add --transport http onlymcp https://only-mcp.com/api/mcp
What are the best alternatives to OnlyMCP?
The closest alternatives to OnlyMCP are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.
Alternatives to OnlyMCP

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

Nautilus Compass Memory
MCP serverLocal-first long-term memory layer for AI agents, served over MCP
Not checked yet- 5 tools
- Remote
- Open source
- 1.3K GitHub stars
- Knowledge and memory

note1
MCP serverMCP server giving AI assistants search and summaries across your note1 meetings
- 4 tools
- Remote
- Free trial
- 1 GitHub star
- Productivity
MedXpert 图书馆连接器(通用版)
MCP serverLocal MCP server that searches and answers questions over your own Markdown wiki
Runs locally- 4 tools
- Open source
- Knowledge and memory