Merge pull request #70 from cudeso/master

Submit malware samples

misp_modules/modules/expansion/vmray_submit.py

@@ -6,7 +6,6 @@ Submit sample to  VMRay.
 Submit a sample to VMRay
 
 TODO:
- # Deal with malicious samples (ZIP file, 'infected')
  # Deal with archive submissions
 
 '''
@@ -15,12 +14,13 @@ import json
 import base64
 
 import io
+import zipfile
 
 from ._vmray.vmray_rest_api import VMRayRESTAPI
 
 misperrors = {'error': 'Error'}
-mispattributes = {'input': ['attachment'], 'output': ['text', 'sha1', 'sha256', 'md5', 'link']}
-moduleinfo = {'version': '0.1', 'author': 'Koen Van Impe',
+mispattributes = {'input': ['attachment', 'malware-sample'], 'output': ['text', 'sha1', 'sha256', 'md5', 'link']}
+moduleinfo = {'version': '0.2', 'author': 'Koen Van Impe',
               'description': 'Submit a sample to VMRay',
               'module-type': ['expansion']}
 moduleconfig = ['apikey', 'url', 'shareable', 'do_not_reanalyze', 'do_not_include_vmrayjobids']
@@ -38,8 +38,23 @@ def handler(q=False):
 
     try:
         data = request.get("data")
-        attachment = request.get("attachment")
-        data = base64.b64decode(data)
+        if 'malware-sample' in request:
+            # malicious samples are encrypted with zip (password infected) and then base64 encoded
+            sample_filename = request.get("malware-sample").split("|",1)[0]
+            data = base64.b64decode(data)
+            fl = io.BytesIO(data)
+            zf = zipfile.ZipFile(fl)
+            sample_hashname = zf.namelist()[0]
+            data = zf.read(sample_hashname,b"infected")
+            zf.close()
+        elif 'attachment' in request:
+            # All attachments get base64 encoded
+            sample_filename = request.get("attachment")
+            data = base64.b64decode(data)
+
+        else:
+            misperrors['error'] = "No malware sample or attachment supplied"
+            return misperrors
     except:
         misperrors['error'] = "Unable to process submited sample data"
         return misperrors
@@ -74,10 +89,10 @@ def handler(q=False):
         do_not_include_vmrayjobids = False
     include_vmrayjobids = not do_not_include_vmrayjobids
 
-    if data and attachment:
+    if data and sample_filename:
         args = {}
         args["shareable"] = shareable
-        args["sample_file"] = {'data': io.BytesIO(data), 'filename': attachment}
+        args["sample_file"] = {'data': io.BytesIO(data), 'filename': sample_filename}
         args["reanalyze"] = reanalyze
 
         try:

misp_modules/modules/import_mod/vmray_import.py

@@ -84,6 +84,7 @@ def handler(q=False):
             # Get all information on the sample, returns a set of finished analyze jobs
             data = vmrayGetInfoAnalysis(api, sample_id)
             if data["data"]:
+                vti_patterns_found = False
                 for analysis in data["data"]:
                     analysis_id = analysis["analysis_id"]
 
@@ -93,17 +94,23 @@ def handler(q=False):
 
                         if analysis_data:
                             p = vmrayVtiPatterns(analysis_data["vti_patterns"])
-                            if p:
-                                if include_analysisid:
-                                    url1 = "https://cloud.vmray.com/user/analysis/view?from_sample_id=%u" % sample_id
-                                    url2 = "&id=%u" % analysis_id
-                                    url3 = "&sub=%2Freport%2Foverview.html"
-                                    p["results"].append({"values": url1 + url2 + url3, "types": "link"})
+                            if p and len(p["results"]) > 0:
+                                vti_patterns_found = True
                                 vmray_results = {'results': vmray_results["results"] + p["results"]}
-
+                            if include_analysisid:
+                                a_id = {'results': []}
+                                url1 = "https://cloud.vmray.com/user/analysis/view?from_sample_id=%u" % sample_id
+                                url2 = "&id=%u" % analysis_id
+                                url3 = "&sub=%2Freport%2Foverview.html"
+                                a_id["results"].append({ "values": url1 + url2 + url3, "types": "link" })
+                                vmray_results = {'results': vmray_results["results"] + a_id["results"] }
                 # Clean up (remove doubles)
-                vmray_results = vmrayCleanup(vmray_results)
-                return vmray_results
+                if vti_patterns_found:
+                    vmray_results = vmrayCleanup(vmray_results)
+                    return vmray_results
+                else:
+                    misperrors['error'] = "No vti_results returned or jobs not finished"
+                    return misperrors
             else:
                 misperrors['error'] = "Unable to fetch sample id %u" % (sample_id)
                 return misperrors
@@ -172,6 +179,8 @@ def vmrayVtiPatterns(vti_patterns):
                 content = vmrayGeneric(pattern)
             elif only_network_info is False and pattern["category"] == "_process" and pattern["operation"] == "_install_ipc_endpoint":
                 content = vmrayGeneric(pattern, "mutex", 1)
+            elif only_network_info is False and pattern["category"] == "_process" and pattern["operation"] == "_crashed_process":
+                content = vmrayGeneric(pattern)
 
             elif only_network_info is False and pattern["category"] == "_anti_analysis" and pattern["operation"] == "_delay_execution":
                 content = vmrayGeneric(pattern)
@@ -240,9 +249,11 @@ def vmrayGeneric(el, attr="", attrpos=1):
         if content:
             if attr:
                 content_split = content.split("\"")
-                content_split[attrpos] = vmraySanitizeInput(content_split[attrpos])
-                r["values"].append(content_split[attrpos])
-                r["types"] = [attr]
+                # Attributes are between open " and close "; so use >
+                if len(content_split) > attrpos:
+                    content_split[attrpos] = vmraySanitizeInput(content_split[attrpos])
+                    r["values"].append(content_split[attrpos])
+                    r["types"] = [attr]                
 
             # Adding the value also as text to get the extra description,
             # but this is pretty useless for "url"