Propose a patch

Evidence

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

diff --git a/moodle_mcp/server.py b/moodle_mcp/server.py
--- a/moodle_mcp/server.py
+++ b/moodle_mcp/server.py
@@ -1198,1 +1198,0 @@
-seria um canal de exfiltração (p.ex. `~/.ssh/id_rsa`). Coloca o e-fólio no
diff --git a/tests/test_prompt_injection.py b/tests/test_prompt_injection.py
--- a/tests/test_prompt_injection.py
+++ b/tests/test_prompt_injection.py
@@ -19,1 +19,0 @@
-"<!-- instrução escondida --> Até breve​" + SMUGGLED + "</p>"
@@ -62,1 +61,0 @@
-text = "Até​ breve‮" + SMUGGLED + "\U000e0101fim"
@@ -67,1 +65,0 @@
-"text", ["a < b e c > d", "Avaliação contínua — ½ das notas, 10 m² 👩‍🎓", "plain ascii", ""]
@@ -70,1 +67,0 @@
-expected = text.replace("‍", "")  # o ZWJ é Cf e sai; o resto fica igual
diff --git a/tests/test_token_safety.py b/tests/test_token_safety.py
--- a/tests/test_token_safety.py
+++ b/tests/test_token_safety.py
@@ -52,1 +52,0 @@
-"https://user:[email protected]/webservice/pluginfile.php/1/a.pdf",
@@ -222,1 +221,0 @@
-["http://moodle.novo.example", "ftp://x.example", "https://user:[email protected]", "https://x.example/?a=1"],

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.8.2. 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: moodle-obsidian-mcp: References a credential file in the home directory (KillerSkills verdict caution 52)

## 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 52/100**. Summary: static analysis only — behavior not yet observed.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/moodle-obsidian-mcp/0.8.2

## Findings with a location
- `moodle_mcp/server.py:1198` · References a credential file in the home directory — moodle_mcp/server.py:1198 · References a credential file in the home directory
- `tests/test_prompt_injection.py:67` · Hidden zero-width characters detected — tests/test_prompt_injection.py:67 · Hidden zero-width characters detected — may conceal content
- `tests/test_prompt_injection.py:19` · Hidden zero-width characters detected — tests/test_prompt_injection.py:19 · Hidden zero-width characters detected — may conceal content
- `tests/test_prompt_injection.py:70` · Hidden zero-width characters detected — tests/test_prompt_injection.py:70 · Hidden zero-width characters detected — may conceal content
- `tests/test_prompt_injection.py:62` · Hidden zero-width characters detected — tests/test_prompt_injection.py:62 · Hidden zero-width characters detected — may conceal content
- `tests/test_token_safety.py:52` · URL with embedded credentials — tests/test_token_safety.py:52 · URL with embedded credentials — may be phishing or credential leaking
- `tests/test_token_safety.py:222` · URL with embedded credentials — tests/test_token_safety.py:222 · URL with embedded credentials — may be phishing or credential leaking

## Proposed change
The attached patch removes the 7 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._