Propose a patch

Evidence

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

diff --git a/tests/test_probe_guidance.py b/tests/test_probe_guidance.py
--- a/tests/test_probe_guidance.py
+++ b/tests/test_probe_guidance.py
@@ -209,1 +209,0 @@
-("https://alice:can'[email protected]/a/SKILL.md", "https://skills.example/a/SKILL.md"),
@@ -210,1 +209,0 @@
-("see 'https://u:[email protected]/x' here", "see 'https://host.example/x' here"),
@@ -234,1 +232,0 @@
-"nested": {"url": "HTTPS://user:[email protected]/mcp?scope=read&token=Token123#section"},
@@ -522,1 +519,0 @@
-auditor.audit_data.protocol_version = "HTTPS://user:[email protected]/?token=Token123"
@@ -571,1 +567,0 @@
-data.probes[rule.probe_id].details[key] = ["HTTPS://user:Secret123@host/path?token=Token123", "工具🔎" * 50]
diff --git a/tests/test_redirects.py b/tests/test_redirects.py
--- a/tests/test_redirects.py
+++ b/tests/test_redirects.py
@@ -134,1 +134,0 @@
-with_credentials = unfollowed_redirect(_redirect(303, "https://u:[email protected]/mcp/"))
@@ -139,1 +138,0 @@
-assert unfollowed_redirect(_redirect(307, "https://user:[email protected]/mcp")) == RefusedRedirect(
@@ -147,1 +145,0 @@
-_redirect(303, "/mcp/?token=s3cret#frag", url="https://user:[email protected]/mcp")
@@ -156,1 +153,0 @@
-endpoint = "https://user:[email protected]/mcp"
@@ -159,1 +155,0 @@
-assert policy_refusal(_redirect(308, "https://user:[email protected]/mcp/", url=endpoint)) is None
diff --git a/mcpscore/smoke_network.py b/mcpscore/smoke_network.py
--- a/mcpscore/smoke_network.py
+++ b/mcpscore/smoke_network.py
@@ -126,1 +126,0 @@
-("userinfo_confusion", "http://127.0.0.1:{allowed}@127.0.0.1:{denied}/v/userinfo_confusion"),
@@ -128,1 +127,0 @@
-("backslash_confusion", "http://127.0.0.1:{denied}\\@127.0.0.1:{allowed}/v/backslash_confusion"),
@@ -129,1 +127,0 @@
-("fragment_confusion", "http://127.0.0.1:{denied}/v/fragment_confusion#@127.0.0.1:{allowed}"),
diff --git a/tests/test_skills.py b/tests/test_skills.py
--- a/tests/test_skills.py
+++ b/tests/test_skills.py
@@ -1028,1 +1028,0 @@
-odd["uri"] = f"https://user:{secret}@skills.example/a/SKILL.md"
@@ -1029,1 +1028,0 @@
-odd["frontmatter"]["allowed-tools"] = f"Fetch(https://user:{secret}@api.example/)"
@@ -1030,1 +1028,0 @@
-odd["resources"].append({"uri": f"https://user:{secret}@skills.example/a/scripts/run.py", "size": 1})
diff --git a/tests/test_sarif.py b/tests/test_sarif.py
--- a/tests/test_sarif.py
+++ b/tests/test_sarif.py
@@ -225,1 +225,0 @@
-("https://user:[email protected]/mcp?sig=abc", "mcp.example.com/mcp"),
@@ -273,1 +272,0 @@
-partial_reason="gated at https://user:[email protected]/x?sig=1",
@@ -303,1 +301,0 @@
-("https://user:[email protected]:8443/mcp?sig=abc123#frag", "https://mcp.example.com:8443/mcp"),
@@ -330,1 +327,0 @@
-target = "https://user:[email protected]:8443/mcp?sig=abc123#frag"
@@ -359,1 +355,0 @@
-"❌ The challenge's resource_metadata 'https://user:[email protected]/.well-known/x?sig=abc123#f' "
diff --git a/tests/test_skills_rules.py b/tests/test_skills_rules.py
--- a/tests/test_skills_rules.py
+++ b/tests/test_skills_rules.py
@@ -1481,1 +1481,0 @@
-secret_root = f"https://user:{password}@skills.example/acme/code-review"
@@ -1483,1 +1482,0 @@
-unknown = f"https://user:{password}@skills.example/acme/mcpscore-no-such-skill/SKILL.md"
diff --git a/tests/test_tls.py b/tests/test_tls.py
--- a/tests/test_tls.py
+++ b/tests/test_tls.py
@@ -488,1 +488,0 @@
-"https://user:[email protected]:8443",
diff --git a/tests/test_smoke_network.py b/tests/test_smoke_network.py
--- a/tests/test_smoke_network.py
+++ b/tests/test_smoke_network.py
@@ -276,1 +276,0 @@
-template = "http://127.0.0.1:{allowed}@127.0.0.1:{denied}/"
@@ -277,1 +276,0 @@
-assert fixture.url(template) == f"http://127.0.0.1:{fixture.allowed_port}@127.0.0.1:{fixture.denied_port}/"
@@ -278,1 +276,0 @@
-assert fixture.display(template) == "http://127.0.0.1:<allowed-port>@127.0.0.1:<denied-port>/"

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.32.0. 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: mcpscore: URL with embedded credentials (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/mcpscore/1.32.0

## Findings with a location
- `tests/test_probe_guidance.py:522` · URL with embedded credentials — tests/test_probe_guidance.py:522 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_redirects.py:147` · URL with embedded credentials — tests/test_redirects.py:147 · URL with embedded credentials — may be phishing or credential leaking
- `mcpscore/smoke_network.py:128` · URL with embedded credentials — mcpscore/smoke_network.py:128 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_redirects.py:134` · URL with embedded credentials — tests/test_redirects.py:134 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_redirects.py:139` · URL with embedded credentials — tests/test_redirects.py:139 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_probe_guidance.py:234` · URL with embedded credentials — tests/test_probe_guidance.py:234 · URL with embedded credentials — may be phishing or credential leaking
- `mcpscore/smoke_network.py:126` · URL with embedded credentials — mcpscore/smoke_network.py:126 · URL with embedded credentials — may be phishing or credential leaking
- `mcpscore/smoke_network.py:129` · URL with embedded credentials — mcpscore/smoke_network.py:129 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_probe_guidance.py:571` · URL with embedded credentials — tests/test_probe_guidance.py:571 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_probe_guidance.py:209` · URL with embedded credentials — tests/test_probe_guidance.py:209 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_redirects.py:156` · URL with embedded credentials — tests/test_redirects.py:156 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_redirects.py:159` · URL with embedded credentials — tests/test_redirects.py:159 · URL with embedded credentials — may be phishing or credential leaking

## Proposed change
The attached patch removes the 27 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._