Model Context Protocol - Revision history

Tom: /* Context */

2026-01-23T22:34:50Z

‎Context

2026-01-23T22:34:06Z

‎Context

2026-01-23T22:33:51Z

‎Context

2026-01-07T01:06:34Z

‎Solutions

2025-11-29T01:19:44Z

‎Context

2025-11-28T23:47:10Z

‎Context

2025-11-28T23:41:57Z

‎Context

2025-11-28T21:17:00Z

‎Context

2025-09-11T22:57:19Z

‎Solutions

2025-09-09T22:48:30Z

‎Authorization

@@ Line 4: / Line 4: @@
 ==Context==
 For a description of what the context is that the AI wants to hoover see the wiki [[AI Context]]
+==The Protocol==
 [https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25 release from Anthropic.
 First, an MCP client will connect to an MCP server with an “initialize” message and negotiate various protocol settings. Second, the server assigns a session ID (like Mcp-Session-Id) that can be used to coordinate all subsequent interactions for that client. This session maintains critical context/state, including:

@@ Line 4: / Line 4: @@
 ==Context==
 For a description of what the context is that the AI wants to hoover see the wiki [[AI Context]]
 [https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25 release from Anthropic.

@@ Line 3: / Line 3: @@
 ==Context==
 [https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25 release from Anthropic.
 First, an MCP client will connect to an MCP server with an “initialize” message and negotiate various protocol settings. Second, the server assigns a session ID (like Mcp-Session-Id) that can be used to coordinate all subsequent interactions for that client. This session maintains critical context/state, including:

@@ Line 130: / Line 130: @@
 * 2025-09 [https://github.com/decentralized-identity/org/blob/main/Org%20documents/WG%20documents/DIF_Trusted_AI_Agents_WG_charter_v1.pdf DIF Trust AI Agents WG charter]
 * 2025-08 [https://github.com/modelcontextprotocol/modelcontextprotocol/issues/1299 SEP-1299: Server-Side Authorization Management with Client Session Binding #1299] Dick Hardt
-* See wiki [[Attention Economy]] for more on Agent to Agent(A2A) solutions
+* See wiki [[Attention Economy]] for more on [https://a2a-protocol.org/latest/tutorials/python/3-agent-skills-and-card/ Agent to Agent(A2A) solutions]
 * 2025-05-23  [https://medium.com/kagenti-the-agentic-platform/security-in-and-around-mcp-part-2-mcp-in-deployment-65bdd0ba9dc6 Security In and Around MCP: Part 2 — MCP in Deployment] Maia Iyer, Mariusz Sabath, Alan Cha, Hai Huang, Paolo Dettori, Jarek Cwiklik
 * 2025-05-16 AWS Load MCP from the command line with one call [https://docs.aws.amazon.com/amazonq/latest/qdeveloper-ug/command-line.html Using Amazon Q Developer on the command line]

@@ Line 3: / Line 3: @@
 ==Context==
-[https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25
+[https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25 release from Anthropic.
-\First, an MCP client will connect to an MCP server with an “initialize” message and negotiate various protocol settings. Second, the server assigns a session ID (like Mcp-Session-Id) that can be used to coordinate all subsequent interactions for that client. This session maintains critical context/state, including:
+First, an MCP client will connect to an MCP server with an “initialize” message and negotiate various protocol settings. Second, the server assigns a session ID (like Mcp-Session-Id) that can be used to coordinate all subsequent interactions for that client. This session maintains critical context/state, including:
 Protocol capabilities negotiated between client and server (which optional features are available).

@@ Line 15: / Line 15: @@
   Mcp-Session-Id: session_abc123
   Content-Type: application/json
+  {
    "jsonrpc": "2.0",
@@ Line 29: / Line 28: @@
   Mcp-Session-Id: session_abc123
   Content-Type: application/json
+  {
    "jsonrpc": "2.0",

@@ Line 3: / Line 3: @@
 ==Context==
-Quantum computers are expected to break traditional cryptographic systems like RSA and ECC. In response,
+[https://www.anthropic.com/news/model-context-protocol?ref=blog.identity.foundation Introducing the Model Context Protocol] 2024--11-25
-Don’t get me wrong — [https://www.wexa.ai/ at Wexa,] we see Model Context Protocol as the HTTP of the GenAI era. It’s the missing link to make AI assistants truly agentic and interoperable.
+MCP is a security nightmare! — [https://www.wexa.ai/ at Wexa,] we see Model Context Protocol as the HTTP of the GenAI era. It’s the missing link to make AI assistants truly agentic and interoperable.
 ==Problems==

@@ Line 122: / Line 122: @@
 ===Authorization===
 MCP is the control plane for AI agents. These systems plan, chain tools, and follow paths you didn't design. Identity security here isn't a one-time check. It's continuous verification of who the human is, who the agent is, and exactly what each step is allowed to do.<ref>Prithvi Poreddy. ''Identity Security for MCP Agents: A Four-Layer Continuous Authorization Model'' https://www.linkedin.com/pulse/identity-security-mcp-agents-four-layer-continuous-model-poreddy-mpnvc/?trackingId=7azamgBErYj3Mhz4RdPItQ%3D%3D</ref>
 Aaron Parecki works for a company that sells OAuth [[Identity and Access Management]] and so is incentivized to [https://aaronparecki.com/2025/04/03/15/oauth-for-model-context-protocol Let's fix OAuth in MCP]