Propose a patch

Evidence

The sandbox observed nothing beyond the static findings; the proposal cites those.

diff --git a/kubectl_mcp_tool/k8s_config.py b/kubectl_mcp_tool/k8s_config.py
--- a/kubectl_mcp_tool/k8s_config.py
+++ b/kubectl_mcp_tool/k8s_config.py
@@ -149,1 +149,0 @@
-kubeconfig_env = os.environ.get('KUBECONFIG', '~/.kube/config')
@@ -248,1 +247,0 @@
-kubeconfig_path = os.environ.get('KUBECONFIG', '~/.kube/config')
@@ -356,1 +354,0 @@
-kubeconfig_path = os.environ.get('KUBECONFIG', '~/.kube/config')
@@ -501,1 +498,0 @@
-kubeconfig_path = os.environ.get('KUBECONFIG', '~/.kube/config')
@@ -531,1 +527,0 @@
-kubeconfig_path = os.environ.get('KUBECONFIG', '~/.kube/config')
diff --git a/kubectl_mcp_tool/providers.py b/kubectl_mcp_tool/providers.py
--- a/kubectl_mcp_tool/providers.py
+++ b/kubectl_mcp_tool/providers.py
@@ -56,1 +56,0 @@
-os.environ.get("KUBECONFIG", "~/.kube/config")
diff --git a/kubectl_mcp_tool/mcp_server.py b/kubectl_mcp_tool/mcp_server.py
--- a/kubectl_mcp_tool/mcp_server.py
+++ b/kubectl_mcp_tool/mcp_server.py
@@ -360,1 +360,0 @@
-logger.debug(f"KUBECONFIG: {os.path.expanduser(os.environ.get('KUBECONFIG', '~/.kube/config'))}")
diff --git a/kubectl_mcp_tool/tools/helm.py b/kubectl_mcp_tool/tools/helm.py
--- a/kubectl_mcp_tool/tools/helm.py
+++ b/kubectl_mcp_tool/tools/helm.py
@@ -700,1 +700,0 @@
-pass_credentials: Pass credentials to all domains

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 2.0.0. If the behavior is intended, declare it (host under permissions.network, a line in the README) instead of removing it; the next crawl scores declarations.

Issue / PR body

Title: kubectl-mcp-server: References a credential file in the home directory (KillerSkills verdict caution 71)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.5) and scored it **CAUTION 71/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/kubectl-mcp-server/2.0.0

## Findings with a location
- `kubectl_mcp_tool/k8s_config.py:149` · References a credential file in the home directory — kubectl_mcp_tool/k8s_config.py:149 · References a credential file in the home directory
- `kubectl_mcp_tool/k8s_config.py:248` · References a credential file in the home directory — kubectl_mcp_tool/k8s_config.py:248 · References a credential file in the home directory
- `kubectl_mcp_tool/providers.py:56` · References a credential file in the home directory — kubectl_mcp_tool/providers.py:56 · References a credential file in the home directory
- `kubectl_mcp_tool/k8s_config.py:501` · References a credential file in the home directory — kubectl_mcp_tool/k8s_config.py:501 · References a credential file in the home directory
- `kubectl_mcp_tool/k8s_config.py:531` · References a credential file in the home directory — kubectl_mcp_tool/k8s_config.py:531 · References a credential file in the home directory
- `kubectl_mcp_tool/mcp_server.py:360` · References a credential file in the home directory — kubectl_mcp_tool/mcp_server.py:360 · References a credential file in the home directory
- `kubectl_mcp_tool/tools/helm.py:700` · Credential pass-through configuration — kubectl_mcp_tool/tools/helm.py:700 · Credential pass-through configuration — agent relays sensitive credentials
- `kubectl_mcp_tool/k8s_config.py:356` · References a credential file in the home directory — kubectl_mcp_tool/k8s_config.py:356 · References a credential file in the home directory

## Proposed change
The attached patch removes the 8 lines above. If the behavior is intended, the fix is to declare it instead: list the host under `permissions.network` (or the MCP manifest) and say in the README why the package needs it; the next crawl will re-score it with the declaration taken into account.

## How this will be verified
Nothing is "verified safe" by assertion. When a new version is published, KillerSkills re-runs it in the honeypot and the verdict page updates; you can trigger it early from the verdict page ("Re-scan after I publish a fix"). Reply here or at [email protected] if you believe the verdict is wrong; disputes are reviewed by a person and the outcome is published on the verdict page.

_Proposed through killerskills.ai/threats by a user of the package, under their own GitHub account. The evidence is public; the package was never modified by KillerSkills._