Editorial Policy
How we decide what to publish on MCPserver.in, where our information comes from, and what we do when we get something wrong.
What we publish
Docs, glossary entries, and server-integration guides on this site are written against the real interfaces in our own codebase and against primary sources: the official Model Context Protocol specification, the npm packages we reference by name and version, and, for India-compliance content, the actual text of the DPDP Act and RBI circulars we cite. Code samples are written to compile against the package versions stated in the page, not left as untested pseudocode.
How we review servers
We check that each listed server has a working repository, clear documentation, and is actively maintained. We also test basic functionality to ensure it meets our quality standards. This is a practical check, not a formal security audit — always review a server's source and required permissions yourself before connecting it to sensitive data or production systems.
What we don't do
- We don't invent adoption numbers, star counts, or uptime figures.
- We don't claim a page was "recently updated" unless its content actually changed.
- We don't list a server in the directory without checking its real repository and auth model.
- We're not affiliated with Anthropic, the MCP maintainers, or any cloud provider named in our guides.
Corrections
If you spot a factual error, a broken link, or an outdated code sample, open an issue or pull request against our GitHub repository. Fixes are reviewed against the same sourcing standard as new content before they merge.
Tooling disclosure
We use AI-assisted tooling as part of our writing and validation workflow, alongside an automated schema audit that checks every content entry against its TypeScript interface before it ships. Every page is still reviewed for factual accuracy before it goes live.