Propose a patch

Evidence

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

diff --git a/tests/test_audit.py b/tests/test_audit.py
--- a/tests/test_audit.py
+++ b/tests/test_audit.py
@@ -84,1 +84,0 @@
-return Settings(url=f"https://gateway:{CANARY_PASSWORD}@{CANARY_HOST}", db="auditdb",
diff --git a/src/odoo_mcp/install_check.py b/src/odoo_mcp/install_check.py
--- a/src/odoo_mcp/install_check.py
+++ b/src/odoo_mcp/install_check.py
@@ -363,1 +363,0 @@
-# With an "@" the split itself is unsafe: in https://admin:4242#pw@host the host
diff --git a/src/odoo_mcp/transport/jsonrpc.py b/src/odoo_mcp/transport/jsonrpc.py
--- a/src/odoo_mcp/transport/jsonrpc.py
+++ b/src/odoo_mcp/transport/jsonrpc.py
@@ -64,1 +64,0 @@
-``https://user:pass@host`` is valid in ODOO_URL (HTTP basic auth in front of
diff --git a/src/odoo_mcp/config.py b/src/odoo_mcp/config.py
--- a/src/odoo_mcp/config.py
+++ b/src/odoo_mcp/config.py
@@ -35,1 +35,0 @@
-"""``https://user:pass@host`` -> ``https://host`` for logs; never raises.
diff --git a/tests/policy_fakes.py b/tests/policy_fakes.py
--- a/tests/policy_fakes.py
+++ b/tests/policy_fakes.py
@@ -171,1 +171,0 @@
-self.creds = SimpleNamespace(url="https://gateway:[email protected]", db="db",
diff --git a/tests/test_json2_transport.py b/tests/test_json2_transport.py
--- a/tests/test_json2_transport.py
+++ b/tests/test_json2_transport.py
@@ -249,1 +249,0 @@
-url = fake.url.replace("http://", "http://gw:PW-NOT-LEAKED@")
@@ -320,1 +319,0 @@
-url = fake.url.replace("http://", "http://gw:secret@")
diff --git a/tests/test_install_check.py b/tests/test_install_check.py
--- a/tests/test_install_check.py
+++ b/tests/test_install_check.py
@@ -374,1 +374,0 @@
-report = check_options(_options(odoo_url=f"https://bot:{SECRET}@my-company.odoo.com"))
@@ -384,1 +383,0 @@
-f"https://admin:{SECRET}@erp.example.com",
@@ -404,1 +402,0 @@
-url = f"https://admin:4242#{SECRET}@erp.example.com"
diff --git a/tests/test_diagnose_cli.py b/tests/test_diagnose_cli.py
--- a/tests/test_diagnose_cli.py
+++ b/tests/test_diagnose_cli.py
@@ -147,1 +147,0 @@
-secret = "https://admin:[email protected]"
diff --git a/tests/test_diagnostics.py b/tests/test_diagnostics.py
--- a/tests/test_diagnostics.py
+++ b/tests/test_diagnostics.py
@@ -196,1 +196,0 @@
-monkeypatch, tmp_path, f"http://{CANARY['user']}:{CANARY['url_password']}@{host}")

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.6.3. 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: odoo-mcp-tools: URL with embedded credentials (KillerSkills verdict caution 76)

## 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 76/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/odoo-mcp-tools/1.6.3

## Findings with a location
- `tests/test_audit.py:84` · URL with embedded credentials — tests/test_audit.py:84 · URL with embedded credentials — may be phishing or credential leaking
- `src/odoo_mcp/install_check.py:363` · URL with embedded credentials — src/odoo_mcp/install_check.py:363 · URL with embedded credentials — may be phishing or credential leaking
- `src/odoo_mcp/transport/jsonrpc.py:64` · URL with embedded credentials — src/odoo_mcp/transport/jsonrpc.py:64 · URL with embedded credentials — may be phishing or credential leaking
- `src/odoo_mcp/config.py:35` · URL with embedded credentials — src/odoo_mcp/config.py:35 · URL with embedded credentials — may be phishing or credential leaking
- `tests/policy_fakes.py:171` · URL with embedded credentials — tests/policy_fakes.py:171 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_json2_transport.py:249` · URL with embedded credentials — tests/test_json2_transport.py:249 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_json2_transport.py:320` · URL with embedded credentials — tests/test_json2_transport.py:320 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_install_check.py:374` · URL with embedded credentials — tests/test_install_check.py:374 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_diagnose_cli.py:147` · URL with embedded credentials — tests/test_diagnose_cli.py:147 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_install_check.py:404` · URL with embedded credentials — tests/test_install_check.py:404 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_install_check.py:384` · URL with embedded credentials — tests/test_install_check.py:384 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_diagnostics.py:196` · URL with embedded credentials — tests/test_diagnostics.py:196 · URL with embedded credentials — may be phishing or credential leaking

## Proposed change
The attached patch removes the 12 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._