Skim MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 94 installs a week from pypi. Last commit 7 Oct 2026.
Lossless skim-then-expand reading of big files, repos, and command output with far fewer tokens
We read the source, 2 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.
proc = subprocess.Popen(command, **kwargs)
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([sys.executable, "-m", "pytest", "--collect-only", "-q"],
r = skim_file(os.path.join(_EVAL, fname))
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 skim-mcp -- uvx skim-mcp
{
"mcpServers": {
"skim-mcp": {
"command": "uvx",
"args": [
"skim-mcp"
]
}
}
}
[mcp_servers.skim-mcp]
command = "uvx"
args = ["skim-mcp"]
{
"mcpServers": {
"skim-mcp": {
"command": "uvx",
"args": [
"skim-mcp"
]
}
}
}
{
"servers": {
"skim-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"skim-mcp"
]
}
}
}
Compress and decompress files with gzip and report the compression ratio.
Serves coding agents one canonical doc per query instead of rereading the repo — ~60% fewer tokens.
Reads a raw DNA file on your computer and reports the variants DeepDNA curates, with sources.
Deterministic, read-only audits of repo readiness for AI coding agents and of agent context files.
Read and post to The Agent Billboard on Solana, with spend limits in code and a reasoning log.
Find when a public company reports: fiscal year end, cadence, next date and the outreach window.
Reading order for big pull requests: what to read first, what can wait. Offline, no model.
Index your codebase. AI searches instead of re-reading files. 94% token savings.
Answers built from our own checks of this server.