ExtractMarkdownLABSCO SUMMARY
It is the pile of small chores you would otherwise write for the fourth time.
None of these calls is hard to build; all of them are tedious to build again. Cleaning a page to Markdown, repairing JSON a model produced, stripping personal data before something is logged — each is one call with nothing to wire up.
What it is
A collection of single-purpose micro-API tools in one .NET server, run locally from a checkout of the repository.
What you get
- ExtractMarkdown — any URL scraped and converted to clean Markdown, with nav, footer and scripts stripped to keep the token count down
- RepairJson — malformed JSON repaired
- InferSchema — the format of a JSON, CSV or XML payload detected and its schema inferred: field names, types, nullability and nesting
- RedactPii — personal data removed from a payload; CleanCsv tidies a CSV
- VerifyEmail — an address checked for disposable and burner domains, by format, against a curated list, and by MX record; GradeEmailAuth grades a domain's email authentication
- ValidatePhone and ValidateTaxId — a phone number and a tax identifier validated
- ResolveTimezone — latitude and longitude turned into IANA timezone, UTC offset, DST status, current local time and next transition
- CountTokens — a token count for a piece of text
- GenerateProductSchema — JSON-LD Product structured data for SEO, returned both raw and as a ready-to-embed script tag
Requirements
The .NET runtime and a clone of the repository — the client runs it with dotnet against the project directory. No API keys or request plumbing on your side.
Setup effort
Build from source — git clone https://github.com/Benzene66/atomicapis-mcp-server.git
USE CASES
01
Getting a web page into the conversation without the clutter
Scrape it to Markdown with the navigation and scripts removed, and check what it costs before pasting it in.
02
Fixing JSON a model produced
Repair the payload, then infer its schema to confirm the fields are the ones you expected.
03
Keeping personal data out of what you log
Redact the payload before it is written anywhere, and tidy the CSV it came from.
04
Checking a signup is a real person
Test the address against burner domains and MX records, validate the phone number, and grade the domain's mail setup.
05
Marking up a product page for search engines
Generate the Product structured data and take the script tag straight into the template.
17 tools
Listed in project docs. Your client may show a different list.
VerifyEmailResolveTimezoneInferSchemaGenerateProductSchemaCleanCsvCloakLinkRepairJsonCountTokensGradeEmailAuthRedactPiiValidatePhoneValidateTaxIdDetectPromptInjectionSemanticSearchFindRelatedContentParseTasksBEFORE YOU INSTALL
What to know first
Tools you get
17 tools named in the project's documentation.
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
Clone the repository.
TERMINAL
git clone https://github.com/Benzene66/atomicapis-mcp-server.gitInstall the .NET 9 SDK; the client configuration runs the project with dotnet.
In the client configuration, <YOUR_PATH_TO_ATOMICAPIS_MCP_SERVER>/src is the absolute path of the src folder in your clone.
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"atomicapis": {
"command": "dotnet",
"args": [
"run",
"--project",
"<YOUR_PATH_TO_ATOMICAPIS_MCP_SERVER>/src"
]
}
}
}Claude Code
CLAUDE CODE · TERMINAL
claude mcp add atomicapis -- dotnet run --project <YOUR_PATH_TO_ATOMICAPIS_MCP_SERVER>/src
