Fix claude md

CLAUDE.md

@@ -8,14 +8,14 @@ See the [design.md](../docs/design.md) for an overview of kperf's architecture a
 
 ## Build
 
-See the [build.md](../docs/build.md) for build instructions to compile and install kperf and runkperf.
+See the [build.md](./docs/build.md) for build instructions to compile and install kperf and runkperf.
 
 ## Setup
 
-See [setup.md](../docs/setup.md) for setup instructions on how to prepare your environment for running kperf and runkperf.
+See [setup.md](./docs/setup.md) for setup instructions on how to prepare your environment for running kperf and runkperf.
 
 ## Test
 
-See the [kperf.md](../docs/kperf.md) for instructions on how to test kperf end-to-end.
+See the [kperf.md](./docs/kperf.md) for instructions on how to test kperf end-to-end.
 
-See the [runkperf.md](../docs/runkperf.md) for instructions how to test runkperf end-to-end.
+See the [runkperf.md](./docs/runkperf.md) for instructions how to test runkperf end-to-end.

docs/kperf.md

@@ -183,4 +183,4 @@ kperf vc nodepool delete example
 - Runner groups use Helm releases deployed in the `runnergroups-kperf-io` namespace
 - Virtual nodes are managed in the `virtualnodes-kperf-io` namespace
 - By default, job controller pods complete after 5 seconds; other pods run until deleted
-- Only one long-running server is allowed per cluster currently
+- Only one long-running server is allowed per cluster currently

docs/setup.md

@@ -89,4 +89,4 @@ eksctl utils write-kubeconfig --cluster=<CLUSTER_NAME> --region=<REGION>
 To configure kubectl to connect to your GKE cluster:
 ```bash
 gcloud container clusters get-credentials <CLUSTER_NAME> --region <REGION>
-```
+```