From ca927a08e144a789441545fc31911fee70e239d7 Mon Sep 17 00:00:00 2001
From: Lucas Bourtoule <dhalf@lucass-mbp.home>
Date: Fri, 13 Oct 2023 16:55:54 +0200
Subject: [PATCH] msgpack-numpy rule

---
 python/msgpack-numpy.py   | 21 ++++++++++++++++++++
 python/msgpack-numpy.yaml | 40 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 61 insertions(+)
 create mode 100644 python/msgpack-numpy.py
 create mode 100644 python/msgpack-numpy.yaml

diff --git a/python/msgpack-numpy.py b/python/msgpack-numpy.py
new file mode 100644
index 0000000..b7b3923
--- /dev/null
+++ b/python/msgpack-numpy.py
@@ -0,0 +1,21 @@
+import msgpack
+import msgpack_numpy as m
+import numpy as np
+
+x = np.random.rand(5)
+# ruleid: msgpack-numpy
+x_enc = msgpack.packb(x, default=m.encode)
+# ruleid: msgpack-numpy
+x_rec = msgpack.unpackb(x_enc, object_hook=m.decode)
+
+# ok: msgpack-numpy
+x_enc2 = msgpack.packb(x)
+# ok: msgpack-numpy
+x_rec2 = msgpack.unpackb(x_enc2)
+
+m.patch()
+
+# ruleid: msgpack-numpy
+x_enc3 = msgpack.packb(x)
+# ruleid: msgpack-numpy
+x_rec3 = msgpack.unpackb(x_enc2)
diff --git a/python/msgpack-numpy.yaml b/python/msgpack-numpy.yaml
new file mode 100644
index 0000000..91b9ba1
--- /dev/null
+++ b/python/msgpack-numpy.yaml
@@ -0,0 +1,40 @@
+rules:
+- id: msgpack-numpy
+  message: >-
+    Found usage of msgpack-numpy unpacking, which relies on pickle to deserialize numpy arrays containing objects.
+    Functions reliant on pickle can result in arbitrary code execution. 
+    Consider switching to a safer serialization method.
+  languages: [python]
+  severity: ERROR
+  metadata:
+    category: security
+    cwe: "CWE-502: Deserialization of Untrusted Data"
+    subcategory: [vuln]
+    confidence: MEDIUM
+    likelihood: MEDIUM
+    impact: HIGH
+    technology: [numpy]
+    description: "Potential arbitrary code execution from functions reliant on pickling"
+    references:
+      - https://blog.trailofbits.com/2021/03/15/never-a-dill-moment-exploiting-machine-learning-pickle-files/
+
+  pattern-either:
+    - patterns:
+      - pattern: msgpack.$FN(...)
+      - metavariable-regex:
+          metavariable: $FN
+          regex: (loads?|dumps?|packb?|unpackb?)
+      - pattern-inside: |
+          msgpack_numpy.patch()
+          ...
+    - pattern-either:
+      - patterns:
+        - pattern: msgpack.$FN(..., object_hook=msgpack_numpy.decode, ...)
+        - metavariable-regex:
+            metavariable: $FN
+            regex: unpackb?
+      - patterns:
+        - pattern: msgpack.$FN(..., default=msgpack_numpy.encode, ...)
+        - metavariable-regex:
+            metavariable: $FN
+            regex: packb?