Propose a patch

Evidence

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

diff --git a/src/neo_mcp/integrations/providers/wandb.py b/src/neo_mcp/integrations/providers/wandb.py
--- a/src/neo_mcp/integrations/providers/wandb.py
+++ b/src/neo_mcp/integrations/providers/wandb.py
@@ -1,1 +1,0 @@
-"""Weights & Biases API key — stored via SecretStore + ~/.netrc."""
@@ -25,1 +24,0 @@
-"""Insert or replace the api.wandb.ai block in ~/.netrc."""
diff --git a/src/neo_mcp/integrations/providers/github.py b/src/neo_mcp/integrations/providers/github.py
--- a/src/neo_mcp/integrations/providers/github.py
+++ b/src/neo_mcp/integrations/providers/github.py
@@ -29,1 +29,0 @@
-kept.append(f"https://{username}:{pat}@github.com")
diff --git a/src/neo_mcp/integrations/providers/s3.py b/src/neo_mcp/integrations/providers/s3.py
--- a/src/neo_mcp/integrations/providers/s3.py
+++ b/src/neo_mcp/integrations/providers/s3.py
@@ -1,1 +1,0 @@
-"""AWS S3 credentials — stored via SecretStore + ~/.aws/credentials [neo] profile."""
@@ -26,1 +25,0 @@
-"""Insert or replace the [neo] profile in ~/.aws/credentials."""
@@ -58,1 +56,0 @@
-"""Remove the [neo] profile from ~/.aws/credentials. Returns True if removed."""
diff --git a/tests/test_system.py b/tests/test_system.py
--- a/tests/test_system.py
+++ b/tests/test_system.py
@@ -2286,1 +2286,0 @@
-"https://user:[email protected]\n"
@@ -2681,1 +2680,0 @@
-raise RuntimeError("connection refused from https://user:[email protected]")

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.5.16. 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: neo-mcp: 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 1.8.1) and scored it **CAUTION 68/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/neo-mcp/0.5.16

## Findings with a location
- `src/neo_mcp/integrations/providers/wandb.py:1` · References a credential file in the home directory — src/neo_mcp/integrations/providers/wandb.py:1 · References a credential file in the home directory
- `src/neo_mcp/integrations/providers/github.py:29` · URL with embedded credentials — src/neo_mcp/integrations/providers/github.py:29 · URL with embedded credentials — may be phishing or credential leaking
- `src/neo_mcp/integrations/providers/s3.py:1` · References a credential file in the home directory — src/neo_mcp/integrations/providers/s3.py:1 · References a credential file in the home directory
- `src/neo_mcp/integrations/providers/s3.py:26` · References a credential file in the home directory — src/neo_mcp/integrations/providers/s3.py:26 · References a credential file in the home directory
- `src/neo_mcp/integrations/providers/s3.py:58` · References a credential file in the home directory — src/neo_mcp/integrations/providers/s3.py:58 · References a credential file in the home directory
- `src/neo_mcp/integrations/providers/wandb.py:25` · References a credential file in the home directory — src/neo_mcp/integrations/providers/wandb.py:25 · References a credential file in the home directory
- `tests/test_system.py:2286` · URL with embedded credentials — tests/test_system.py:2286 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_system.py:2681` · URL with embedded credentials — tests/test_system.py:2681 · URL with embedded credentials — may be phishing or credential leaking

## 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._