Delimit — Keep the state. Change the model. runs on your own machine — the client starts it, so there is no endpoint to ping. 1 373 installs a week from npm. Last commit 8 Nov 2026.
Keep the state. Change the model. Shared context and a multi-model panel across coding assistants.
We read the source, 12 h ago · rules 995398eba69f
Things with no honest explanation: a promise that contradicts the code, code that runs at install time while hiding what it does, data leaving the machine.
probe = subprocess.run(
env = dict(os.environ)
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
postinstall: node scripts/postinstall.js
__import__(pkg)
return subprocess.check_output(
self.command, shell=True,
execSync(`${cmd} --version`, { stdio: 'pipe' });
path.join(codexConfig.configPath, codexConfig.configFile),
for child in sorted(p.iterdir()):
'/.ssh/',
env = dict(os.environ)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
Code changes quietly between releases, and nobody reads the diff of a dependency. We do, on every release — watch Delimit — Keep the state. Change the model. and you get told the day something new turns up.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add delimit-mcp-server -- npx -y delimit-cli
{
"mcpServers": {
"delimit-mcp-server": {
"command": "npx",
"args": [
"-y",
"delimit-cli"
]
}
}
}
[mcp_servers.delimit-mcp-server]
command = "npx"
args = ["-y", "delimit-cli"]
{
"mcpServers": {
"delimit-mcp-server": {
"command": "npx",
"args": [
"-y",
"delimit-cli"
]
}
}
}
{
"servers": {
"delimit-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"delimit-cli"
]
}
}
}
Persistent memory and shared context across compatible AI assistants, with account-scoped OAuth.
A Model Context Protocol (MCP) server and app for creating Datawrapper charts using AI assistants
One workspace for every AI coding assistant. Shared tasks, memory, and governance.
Share context, detect conflicts, and message across parallel Claude Code sessions
Connect AI assistants to zeroheight styleguides using the Model Context Protocol
Route coding work across Claude Code, Codex, Cursor, and OpenCode with shared context.
Symbol-aware C# and .NET code context for AI coding assistants.
Multi-agent shared brain across Claude, Codex, Antigravity, Gemini, Copilot, and VS Code.
Answers built from our own checks of this server.