Propose a patch

Evidence

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

diff --git a/src/ghost_hands/bench.py b/src/ghost_hands/bench.py
--- a/src/ghost_hands/bench.py
+++ b/src/ghost_hands/bench.py
@@ -949,1 +949,0 @@
-d = ChromiumDriver(proxy="http://user:[email protected]:3128")
diff --git a/src/ghost_hands/phone_approver.py b/src/ghost_hands/phone_approver.py
--- a/src/ghost_hands/phone_approver.py
+++ b/src/ghost_hands/phone_approver.py
@@ -247,1 +247,0 @@
-"no Ghost Hands pairing token: pass token= or set "
diff --git a/src/ghost_hands/drivers.py b/src/ghost_hands/drivers.py
--- a/src/ghost_hands/drivers.py
+++ b/src/ghost_hands/drivers.py
@@ -17,1 +17,0 @@
-proxy credentials in ``--proxy-server``), and stale-target detection in
diff --git a/src/ghost_hands/android_driver.py b/src/ghost_hands/android_driver.py
--- a/src/ghost_hands/android_driver.py
+++ b/src/ghost_hands/android_driver.py
@@ -171,1 +171,0 @@
-"no Ghost Hands pairing token: pass token= or set "

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.11.0. 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: ghost-hands: URL with embedded credentials (KillerSkills verdict caution 73)

## 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 73/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/glama/ghost-hands/1.11.0

## Findings with a location
- `src/ghost_hands/bench.py:949` · URL with embedded credentials — src/ghost_hands/bench.py:949 · URL with embedded credentials — may be phishing or credential leaking
- `src/ghost_hands/phone_approver.py:247` · Credential pass-through configuration — src/ghost_hands/phone_approver.py:247 · Credential pass-through configuration — agent relays sensitive credentials
- `src/ghost_hands/drivers.py:17` · Credential pass-through configuration — src/ghost_hands/drivers.py:17 · Credential pass-through configuration — agent relays sensitive credentials
- `src/ghost_hands/android_driver.py:171` · Credential pass-through configuration — src/ghost_hands/android_driver.py:171 · Credential pass-through configuration — agent relays sensitive credentials

## 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._