docs(mcp): use ILP solver in walkthrough (default, more realistic)

GiggleLiu · claude · GiggleLiu · commit f0cac97bda1b · 2026-02-23T01:17:21.000+08:00
Co-Authored-By: Claude Opus 4.6 &lt;noreply@anthropic.com&gt;
diff --git a/docs/src/mcp.md b/docs/src/mcp.md
@@ -33,7 +33,7 @@ A complete session showing how an AI assistant uses MCP tools behind the scenes
 
 **You:** Now solve it through the reduction, explain the result.
 
-> **Assistant:** The brute-force solver found the optimal solution by solving the QUBO, then mapping the result back to the original MIS:
+> **Assistant:** The ILP solver found the optimal solution by solving the QUBO, then mapping the result back to the original MIS:
 >
 > - **QUBO solution:** [0,1,0,1] with objective -6.0 (minimization)
 > - **MIS solution:** vertices {1, 3} with weight **6** (maximization)

Original file line number	Diff line number	Diff line change
`@@ -33,7 +33,7 @@ A complete session showing how an AI assistant uses MCP tools behind the scenes`
`33`	`33`
`34`	`34`	`You: Now solve it through the reduction, explain the result.`
`35`	`35`
`36`		`-> Assistant: The brute-force solver found the optimal solution by solving the QUBO, then mapping the result back to the original MIS:`
	`36`	`+> Assistant: The ILP solver found the optimal solution by solving the QUBO, then mapping the result back to the original MIS:`
`37`	`37`	`>`
`38`	`38`	`> - QUBO solution: [0,1,0,1] with objective -6.0 (minimization)`
`39`	`39`	`> - MIS solution: vertices {1, 3} with weight 6 (maximization)`