Propose a patch

Evidence

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

diff --git a/src/plaxis_mcp/worker_runtime.py b/src/plaxis_mcp/worker_runtime.py
--- a/src/plaxis_mcp/worker_runtime.py
+++ b/src/plaxis_mcp/worker_runtime.py
@@ -716,1 +716,0 @@
-"anyone to type the password into the chat. The connect tool takes no "
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -87,1 +87,0 @@
-<summary><b>Source / developer launch only</b> — not a supported deployment path</summary> <br> A clean CPython 3.13 environment can run the module directly wit
@@ -231,1 +230,0 @@
-<summary>Historical PLAXIS scripting paths — worker-only, never on the host <code>PYTHONPATH</code></summary> <br> ``` C:\ProgramData\Seequent\PLAXIS Python Dis
diff --git a/tests/test_error_fidelity.py b/tests/test_error_fidelity.py
--- a/tests/test_error_fidelity.py
+++ b/tests/test_error_fidelity.py
@@ -1497,1 +1497,0 @@
-self.assertIn("Do not ask anyone to type the password", message)

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.3.5. 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: plaxis-mcp: Prompting user to enter sensitive credentials (KillerSkills verdict caution 75)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.8.1) and scored it **CAUTION 75/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/plaxis-mcp/0.3.5

## Findings with a location
- `src/plaxis_mcp/worker_runtime.py:716` · Prompting user to enter sensitive credentials — src/plaxis_mcp/worker_runtime.py:716 · Prompting user to enter sensitive credentials — social engineering
- `README.md:231` · Collapsed <details> block carrying instructions — README.md:231 · Collapsed <details> block carrying instructions
- `README.md:87` · Collapsed <details> block carrying instructions — README.md:87 · Collapsed <details> block carrying instructions
- `tests/test_error_fidelity.py:1497` · Prompting user to enter sensitive credentials — tests/test_error_fidelity.py:1497 · Prompting user to enter sensitive credentials — social engineering

## Proposed change
The attached patch removes the 4 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._