Overllm runs on your own machine — the client starts it, so there is no endpoint to ping. 182 installs a week from pypi. Last commit 5 Jul 2026.
Find the LLM/AI calls you didn't need — where plain code or a regex does the job. No model.
What the code does
We read the source, 2 d ago · rules 995398eba69f
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.
proc = subprocess.run(cmd, capture_output=True, text=True)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
This code can reach further than it looks
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Connect this server
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 overllm -- uvx overllm{
"mcpServers": {
"overllm": {
"command": "uvx",
"args": [
"overllm"
]
}
}
}[mcp_servers.overllm]
command = "uvx"
args = ["overllm"]{
"mcpServers": {
"overllm": {
"command": "uvx",
"args": [
"overllm"
]
}
}
}{
"servers": {
"overllm": {
"type": "stdio",
"command": "uvx",
"args": [
"overllm"
]
}
}
}Alternatives to Overllm
same job, measured the same wayFind retiring LLM calls in code and test the replacement model on real traffic before switching.
Finds the LLM model IDs pinned in your code that vendors have already switched off, or have dated to
Reads the open-weight model IDs in your code and names the licence clause that binds you - 22 rules
Risk-scan a diff, flag AI-generated-code tells, find secrets. 5 of 7 tools need no account.
Records your Claude Code, Codex and Pi history before deletion. Cited-or-abstain recall.
Code review by AI models from different companies, usually two. Shows where they agree and disagree.
Clean messy CSVs: an LLM picks the steps, tested code does the work. Data untouched by the model.
Code-review MCP: review a GitLab MR or GitHub PR against the whole codebase, not just the diff.
Overllm — questions
Answers built from our own checks of this server.