Model Context Protocol specification 2026-07-28, Security Best Practices

Publisher

Model Context Protocol project

URL

https://modelcontextprotocol.io/specification/2026-07-28/basic/security_best_practices

Published

2026-07-28

Updated

unknown

Accessed

2026-09-05

HTTP status

200

License

MIT license; permitted under the specification license

Source notes citing this source

• The 5 message flow diagrams on the Model Context Protocol security best practices page depict consent phishing a skipped consent cookie server side request forgery to a cloud metadata endpoint and script injection leading to arbitrary command execution and none of them concerns tool schemas. in context
• Because MCP is stateless and has no protocol-level sessions, servers that need state mint explicit handles, and servers implementing authorization MUST verify all inbound requests and MUST NOT treat possession of a state handle as authentication. in context
• The Model Context Protocol security best practices page carries no section on tool schema change tool poisoning or rug pull attacks and its attack sections are all authorization shaped. in context

Cite this source record

APA

WarmTransfer. (2026, July 28). Model Context Protocol specification 2026-07-28, Security Best Practices. WarmTransfer. https://warmtransfer.net/knowledge/sources/mcp-spec-2026-07-28-security-best-practices

BibTeX

@misc{warmtransfer-mcp-spec-2026-07-28-security-best-practices,
  title  = {Model Context Protocol specification 2026-07-28, Security Best Practices},
  author = {{WarmTransfer}},
  year   = {2026},
  url    = {https://warmtransfer.net/knowledge/sources/mcp-spec-2026-07-28-security-best-practices},
  note   = {Model Context Protocol project, accessed 2026-09-05}
}