Propose a patch

Evidence

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

diff --git a/src/lex_mcp_local/wrapper_mcp.py b/src/lex_mcp_local/wrapper_mcp.py
--- a/src/lex_mcp_local/wrapper_mcp.py
+++ b/src/lex_mcp_local/wrapper_mcp.py
@@ -228,1 +228,0 @@
-"""Write an AGENTS.md file to the project root for LLM discoverability."""
@@ -261,1 +260,0 @@
-logger.warning("Failed to write AGENTS.md: %s", exc)
@@ -1436,1 +1434,0 @@
-# (https://x-access-token:{token}@github.com/...) so no credential helper is needed.
@@ -3672,1 +3669,0 @@
-remote_url = f"https://x-access-token:{token}@github.com/{owner}/{repo}.git"
@@ -3676,1 +3672,0 @@
-# Write AGENTS.md so any LLM working in this directory knows about the framework
@@ -3979,1 +3974,0 @@
-remote_url = f"https://x-access-token:{token}@github.com/{owner}/{repo}.git"
@@ -4325,1 +4319,0 @@
-remote_url = f"https://x-access-token:{token}@github.com/{owner}/{repo}.git"

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.2.4. 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: lex-mcp-local: URL with embedded credentials (KillerSkills verdict caution 72)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.6) and scored it **CAUTION 72/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/lex-mcp-local/0.2.4

## Findings with a location
- `src/lex_mcp_local/wrapper_mcp.py:3979` · URL with embedded credentials — src/lex_mcp_local/wrapper_mcp.py:3979 · URL with embedded credentials — may be phishing or credential leaking
- `src/lex_mcp_local/wrapper_mcp.py:4325` · URL with embedded credentials — src/lex_mcp_local/wrapper_mcp.py:4325 · URL with embedded credentials — may be phishing or credential leaking
- `src/lex_mcp_local/wrapper_mcp.py:3672` · URL with embedded credentials — src/lex_mcp_local/wrapper_mcp.py:3672 · URL with embedded credentials — may be phishing or credential leaking
- `src/lex_mcp_local/wrapper_mcp.py:1436` · URL with embedded credentials — src/lex_mcp_local/wrapper_mcp.py:1436 · URL with embedded credentials — may be phishing or credential leaking
- `src/lex_mcp_local/wrapper_mcp.py:228` · Instructs the agent to modify its persistent memory or instruction files — src/lex_mcp_local/wrapper_mcp.py:228 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `src/lex_mcp_local/wrapper_mcp.py:261` · Instructs the agent to modify its persistent memory or instruction files — src/lex_mcp_local/wrapper_mcp.py:261 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `src/lex_mcp_local/wrapper_mcp.py:3676` · Instructs the agent to modify its persistent memory or instruction files — src/lex_mcp_local/wrapper_mcp.py:3676 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file

## Proposed change
The attached patch removes the 7 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._