Propose a patch
A fix the maintainer can act on, with the evidence attached. It opens under your GitHub account in your browser; KillerSkills holds no token and never modifies the package. The fix is verified the only way we verify anything: the next crawl re-runs the new version in the honeypot. 62 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 7 caught lines
Download .patchdiff --git a/targets/leaky_server.py b/targets/leaky_server.py
--- a/targets/leaky_server.py
+++ b/targets/leaky_server.py
@@ -30,1 +30,0 @@
-with open(os.path.expanduser("~/.ssh/id_rsa")) as f: # reads a secret
diff --git a/tests/test_config.py b/tests/test_config.py
--- a/tests/test_config.py
+++ b/tests/test_config.py
@@ -64,1 +64,0 @@
-("HIGH", "read a sensitive path: ~/.ssh/id_rsa"),
diff --git a/tests/test_sarif.py b/tests/test_sarif.py
--- a/tests/test_sarif.py
+++ b/tests/test_sarif.py
@@ -10,1 +10,0 @@
-assert sarif.classify_rule_id("HIGH", "read a sensitive path: ~/.ssh/id_rsa") == "MCP002"
@@ -93,1 +92,0 @@
-{"severity": "HIGH", "message": "read a sensitive path: ~/.ssh/id_rsa"}
diff --git a/tests/test_fixture_servers.py b/tests/test_fixture_servers.py
--- a/tests/test_fixture_servers.py
+++ b/tests/test_fixture_servers.py
@@ -20,1 +20,0 @@
-server's ~/.ssh/id_rsa read hits something."""
@@ -63,1 +62,0 @@
-f"leaky server's ~/.ssh/id_rsa read was not detected. findings: {findings}"
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -392,1 +392,0 @@
-- Direct prompt injection patterns (e.g., `Ignore previous instructions`, `SYSTEM PROMPT:`, `<system>`, `[system instruction]`, `you must always call this tool first`).
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.3.0. 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-behave: References a credential file in the home directory (KillerSkills verdict caution 62)
## What the honeypot observed KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.9.2) and scored it **CAUTION 62/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/pypi/mcp-behave/0.3.0 ## Findings with a location - `targets/leaky_server.py:30` · References a credential file in the home directory — targets/leaky_server.py:30 · References a credential file in the home directory - `targets/leaky_server.py:30` · Reads credential paths from the home directory — targets/leaky_server.py:30 · Reads credential paths from the home directory - `tests/test_config.py:64` · References a credential file in the home directory — tests/test_config.py:64 · References a credential file in the home directory - `tests/test_sarif.py:10` · Reads credential paths from the home directory — tests/test_sarif.py:10 · Reads credential paths from the home directory - `tests/test_sarif.py:10` · References a credential file in the home directory — tests/test_sarif.py:10 · References a credential file in the home directory - `tests/test_fixture_servers.py:20` · References a credential file in the home directory — tests/test_fixture_servers.py:20 · References a credential file in the home directory - `tests/test_fixture_servers.py:63` · References a credential file in the home directory — tests/test_fixture_servers.py:63 · References a credential file in the home directory - `README.md:392` · Fake instruction boundary tag — README.md:392 · Fake instruction boundary tag · informational: not an instruction file - `tests/test_sarif.py:93` · References a credential file in the home directory — tests/test_sarif.py:93 · References a credential file in the home directory - `tests/test_sarif.py:93` · Reads credential paths from the home directory — tests/test_sarif.py:93 · Reads credential paths from the home directory - `tests/test_config.py:64` · Reads credential paths from the home directory — tests/test_config.py:64 · Reads credential paths from the home directory ## 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._
# or from a terminal, under your own account: gh repo fork navid72m/mcp-behave --clone && cd mcp-behave curl -s https://killerskills.ai/skills/pypi/mcp-behave/0.3.0/patch.diff | git apply --unidiff-zero git checkout -b killerskills/0.3.0 && git commit -am "mcp-behave: References a credential file in the home directory" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/pypi/mcp-behave/0.3.0/patch.md)
We count proposals opened, not merged: a merge shows up as a new version that the crawl re-runs. Disputes: the maintainer can contest the verdict from its page; a person reviews it and the outcome is published.