Section Anything runs on your own machine — the client starts it, so there is no endpoint to ping. 388 installs a week from pypi. Last commit 31 Oct 2026.
A PDF's real sections for agents: heading tree, pages and section text. Free mode is local.
We read the source, 6 h ago · rules 995398eba69f
Found in continuous integration, deployment or infrastructure files, or in a neighbouring package of the same monorepo. None of this is installed when you add the server: it describes how the project is built and released. We list it because a leaked key in a build pipeline is still a real problem, but it is not something this server does on your machine.
proc = subprocess.Popen( # noqa: S603 — fixed argv, no shell
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
Today is the operative word: we check Section Anything every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
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 section-anything -- uvx section-anything-mcp
{
"mcpServers": {
"section-anything": {
"command": "uvx",
"args": [
"section-anything-mcp"
]
}
}
}
[mcp_servers.section-anything]
command = "uvx"
args = ["section-anything-mcp"]
{
"mcpServers": {
"section-anything": {
"command": "uvx",
"args": [
"section-anything-mcp"
]
}
}
}
{
"servers": {
"section-anything": {
"type": "stdio",
"command": "uvx",
"args": [
"section-anything-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
OPENAI_API_KEY (Only for the paid quick and standard modes with OpenAI models), ANTHROPIC_API_KEY (Only for the paid modes with a Claude model).
PDF reader for vision LLMs. Auto-detects text corruption and switches to image mode.
Web search, page reading, HTML to PDF, file upload and flight search for AI agents, paid in USDC.
Generate PDF documents locally with text, headings, and pages. No network and no key.
MCP server for extracting text, images, tables, links, annotations, and metadata from PDF files.
Local PDF, image, data and text tools for AI assistants. Files stay on your computer.
Local file tools for agents: documents, PDFs, images, QR codes. Nothing is uploaded.
43 local developer tools for AI agents: PDF, subnet, regex, JSON, conversion, and more.
No-API-key web search, page fetching and document reading across many engines. Local-first.
Answers built from our own checks of this server.