docs: add audit matrix next steps

Signed-off-by: Ho Lim <subhoya@gmail.com>

Author: HOYALIM

docs/inference/model-capability-audit.mdx

@@ -118,3 +118,9 @@ Do not leave placeholder text in a completed row.
 
 - `nemoclaw-blueprint/model-specific-setup/README.md` documents where model-specific setup belongs once an intervention is justified.
 - `docs/inference/tool-calling-reliability` explains the local inference tool-call failure mode that audit rows should classify separately from provider connectivity.
+
+## Next Steps
+
+- [Inference Options](inference-options) for choosing a provider path before adding audit evidence.
+- [Tool-Calling Reliability](tool-calling-reliability) for separating provider connectivity from model tool-use behavior.
+- [Architecture](../reference/architecture) for the model-specific setup registry location.

test/model-capability-audit-doc.test.ts

@@ -65,4 +65,13 @@ describe("model capability audit doc (#3123)", () => {
       "_build/agent-variants/inference/model-capability-audit.hermes.generated.mdx",
     );
   });
+
+  it("keeps next-step links for related inference docs", () => {
+    const markdown = fs.readFileSync(auditDocPath, "utf8");
+
+    expect(markdown).toContain("## Next Steps");
+    expect(markdown).toContain("[Inference Options](inference-options)");
+    expect(markdown).toContain("[Tool-Calling Reliability](tool-calling-reliability)");
+    expect(markdown).toContain("[Architecture](../reference/architecture)");
+  });
 });