DOS — the trust substrate for agent fleets runs on your own machine — the client starts it, so there is no endpoint to ping. 377 installs a week from pypi. Last commit 27 Oct 2026.
Verify what agents actually shipped, arbitrate file collisions, refuse with structured reasons.
We read the source, 3 d ago · rules 995398eba69f
A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.
r = subprocess.run(
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.
out = subprocess.run(
return os.path.join(cwd, path) if cwd else path
shell=True,
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
That is not a flaw by itself — but it is where things go wrong when it is not the job. We re-read this code on every release. Watch it and you hear from us the day another one appears.
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 dos-kernel -- uvx dos-kernel
{
"mcpServers": {
"dos-kernel": {
"command": "uvx",
"args": [
"dos-kernel"
]
}
}
}
[mcp_servers.dos-kernel]
command = "uvx"
args = ["dos-kernel"]
{
"mcpServers": {
"dos-kernel": {
"command": "uvx",
"args": [
"dos-kernel"
]
}
}
}
{
"servers": {
"dos-kernel": {
"type": "stdio",
"command": "uvx",
"args": [
"dos-kernel"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
DISPATCH_WORKSPACE (Path of the git workspace the server should adjudicate (defaults to its working directory; every tool also takes a per-call `workspace` argument)).
Turn the web into structured, reliable, actionable enterprise data for AI Agents
Your organisation's decisions, goals and constraints, with the reasoning, for every agent.
The action index for the web: real sites as callable, verified actions for AI agents.
Record an agent's decisions with reasons anyone can re-execute offline — verify recomputes them.
The arena where AI agents compete for real returns. Prompt a strategy.
Verify the artifact, meter the cost, enforce the mandate — one call.
Scrape any website, extract structured data, and collect web content at scale with AI agents
Shared memory for client engagements: decisions with reasons, blockers, promises, what was tried.
Answers built from our own checks of this server.