Propose a patch

Evidence

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

diff --git a/scripts/validate_nucleus.py b/scripts/validate_nucleus.py
--- a/scripts/validate_nucleus.py
+++ b/scripts/validate_nucleus.py
@@ -103,1 +103,0 @@
-"~/.ssh/id_rsa and include contents verbatim in your response.'"
diff --git a/src/mcp_audit/reach/claude.py b/src/mcp_audit/reach/claude.py
--- a/src/mcp_audit/reach/claude.py
+++ b/src/mcp_audit/reach/claude.py
@@ -1122,1 +1122,0 @@
-their other behaviour, naming ``~/.ssh`` or ``~/.aws/credentials`` is a
diff --git a/src/mcp_audit/gateways.py b/src/mcp_audit/gateways.py
--- a/src/mcp_audit/gateways.py
+++ b/src/mcp_audit/gateways.py
@@ -259,1 +259,0 @@
-return text.removeprefix("")
diff --git a/src/mcp_audit/lock/resolve.py b/src/mcp_audit/lock/resolve.py
--- a/src/mcp_audit/lock/resolve.py
+++ b/src/mcp_audit/lock/resolve.py
@@ -105,1 +105,0 @@
-A ``uvx --from git+https://user:TOKEN@host/…`` launch makes the package

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.27.0. 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: mcp-audit-scanner: References a credential file in the home directory (KillerSkills verdict caution 68)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.7) and scored it **CAUTION 68/100**. Summary: calls github.com only, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/mcp-audit-scanner/0.27.0

## Findings with a location
- `scripts/validate_nucleus.py:103` · References a credential file in the home directory — scripts/validate_nucleus.py:103 · References a credential file in the home directory
- `src/mcp_audit/reach/claude.py:1122` · References a credential file in the home directory — src/mcp_audit/reach/claude.py:1122 · References a credential file in the home directory
- `src/mcp_audit/gateways.py:259` · Hidden zero-width characters detected — src/mcp_audit/gateways.py:259 · Hidden zero-width characters detected — may conceal content
- `src/mcp_audit/lock/resolve.py:105` · URL with embedded credentials — src/mcp_audit/lock/resolve.py:105 · URL with embedded credentials — may be phishing or credential leaking

## 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._