symbolics-mcp
Provides the ability to evaluate Common Lisp and Zetalisp expressions on a Symbolics Genera machine and retrieve the resulting values, standard output, and error messages.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@symbolics-mcpEvaluate (scl:list-all-processes) to see current activity on the machine"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
lisp-mcp
An MCP (Model Context Protocol) server that gives LLMs the ability to evaluate Common Lisp expressions. It provides up to three tools depending on configuration:
eval_host_cl -- always available; evaluates expressions in the MCP server's own SBCL process
eval_swank -- connects to a running Lisp via the Swank protocol (the backend used by SLIME and Sly)
eval_genera -- bridges to a Symbolics Genera machine over TCP
Prerequisites
SBCL on the host machine
Quicklisp installed in SBCL (the MCP server loads
usocketandyasonvia Quicklisp)
MCP configuration
Add the following to your .mcp.json (or the MCP configuration of your client):
{
"mcpServers": {
"lisp": {
"command": "/path/to/lisp-mcp.sh",
"env": {
"SWANK_PORT": "4005",
"GENERA_HOST": "192.168.11.2"
}
}
}
}Only command is required. The environment variables enable the optional tools.
Tools
eval_host_cl
Evaluate a Common Lisp expression in the MCP server's own SBCL process. Always available, no configuration needed. Stdout, stderr, and the return value are captured and returned.
eval_swank
Evaluate a Common Lisp expression on a running Lisp that has a Swank server. This is the most useful tool for development -- it lets the LLM interact with your live Lisp image, load systems, inspect state, and run tests.
The Swank client implements the Swank wire protocol, handles debugger entry by automatically invoking the *ABORT restart and returning the error description, and responds to interactive prompts (e.g. y-or-n-p during compilation).
Start a Swank server in your target Lisp:
(swank:create-server :port 4005 :dont-close t)The :dont-close t argument keeps the server listening after the first connection disconnects, so the MCP server can reconnect between requests.
To have Swank automatically start an additional server on a fixed port whenever you launch SLIME from Emacs with M-x slime, add the following to your Emacs configuration:
(add-hook 'slime-connected-hook
(lambda ()
(slime-eval-async '(swank:create-server :port 4005 :dont-close t))))Then set SWANK_PORT in the MCP configuration to enable the tool.
Variable | Default | Description |
| (unset) | TCP port of the Swank server. When set, enables |
|
| Hostname or IP of the Swank server |
|
| Package in which expressions are evaluated |
eval_genera
Evaluate a Lisp expression on a Symbolics Genera machine. Requires eval-server.lisp running on the Genera side.
Set GENERA_HOST in the MCP configuration to enable the tool.
Variable | Default | Description |
| (unset) | Hostname or IP of the Genera machine. When set, enables |
|
| TCP port of the eval server |
|
| Timeout in seconds for each evaluation |
Setting up the Genera side
Copy eval-server.lisp to the Genera SYS:SITE; directory on the host filesystem:
cp eval-server.lisp /opt/symbolics/lib/sys.sct/site/eval-server.lispThen load it from the Genera Lisp Listener:
Load File SYS:SITE;eval-server.lispThe file must be loaded once per boot. It defines a TCP server (:mcp-server) on port 8888 that accepts connections, reads length-prefixed Lisp expressions, evaluates them, and returns the result, stdout, and stderr as length-prefixed fields.
To verify it is running:
echo -n '11 (+ 1 2 3 4)' | nc -w 2 192.168.11.2 8888 | cat -vThis server cannot be installed
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
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/hanshuebner/genera-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server