Zar-Branch (#6)
Co-authored-by: = <=> Co-authored-by: Driven-Element <129630760+Driven-Element@users.noreply.github.com> Reviewed-on: brotoskyj/AirlockTools#6 Co-authored-by: brotoskyj <jbrotosky@gmail.com> Co-committed-by: brotoskyj <jbrotosky@gmail.com>
This commit was merged in pull request #6.
This commit is contained in:
brotoskyj
committed by
brotoskyj
brotoskyj
co-authored by
=
Driven-Element
=
Driven-Element
parent
4d25a21ddb
commit
ccbf716418
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
@@ -0,0 +1,16 @@
|
||||
import pandas as pd
|
||||
|
||||
def filter_and_drop(approved, eligiblepaths, min_hashes):
|
||||
"""
|
||||
Filters eligiblepaths to rows where all hashes are in approved,
|
||||
then drops rows with fewer than min_hashes hashes.
|
||||
"""
|
||||
approved_hashes = set(approved['sha256'])
|
||||
|
||||
def all_hashes_approved(row):
|
||||
return all(h in approved_hashes for h in row['sha256'])
|
||||
|
||||
filtered = eligiblepaths[eligiblepaths.apply(all_hashes_approved, axis=1)]
|
||||
filtered = filtered[filtered['sha256'].apply(len) >= min_hashes]
|
||||
|
||||
return filtered
|
||||
+58
-24
@@ -2,6 +2,7 @@ import datetime
|
||||
import requests
|
||||
import json
|
||||
import os
|
||||
import time
|
||||
|
||||
def allowlistexechistories(url, outputjson: bool):
|
||||
endpoint = url + '/v1/group'
|
||||
@@ -20,38 +21,71 @@ def allowlistexechistories(url, outputjson: bool):
|
||||
policyids.append(list['groupid'])
|
||||
choice = input("Select Policy Group: ")
|
||||
choice = int(choice) - 1
|
||||
endpoint = url + '/v1/logging/exechistories'
|
||||
payload_dict = {
|
||||
"type":[1, 2, 6, 7],
|
||||
"checkpoint":"68a153c23963989b484541b4",
|
||||
"policy": [policiesnames[choice]]
|
||||
}
|
||||
payload = json.dumps(payload_dict)
|
||||
print(payload)
|
||||
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
|
||||
|
||||
checkpoint = '000000000000000000000000'
|
||||
json_output = {'error': 'Success', 'response': {'exechistories': []}}
|
||||
while True:
|
||||
json_response_data = checkpoint_stomper(checkpoint, url, policiesnames[choice], headers)
|
||||
if not json_response_data['response']['exechistories']:
|
||||
break
|
||||
for index, item in enumerate(json_response_data['response']['exechistories']):
|
||||
if index == len(json_response_data['response']['exechistories']) -1:
|
||||
checkpoint = item['checkpoint']
|
||||
print(f"Date Greater than 30 Days, Stepping to new Checkpoint. {item['checkpoint']}")
|
||||
else:
|
||||
if (datetime.date.today() - datetime.timedelta(days=10) > datetime.datetime.strptime(item['datetime'].replace(' +0000 UTC', ''), '%Y-%m-%dT%H:%M:%SZ').date()):
|
||||
pass
|
||||
else:
|
||||
#json_output['response']['exechistories'].append(json_response_data['response']['exechistories'][1])
|
||||
for output in json_response_data['response']['exechistories']:
|
||||
json_output['response']['exechistories'].append(output)
|
||||
json_output = json.dumps(json_output)
|
||||
if outputjson == True:
|
||||
return response
|
||||
|
||||
parse_text = json.loads(response.text)
|
||||
for item in parse_text['response']['exechistories']:
|
||||
print(item['checkpoint'])
|
||||
print(item['datetime'])
|
||||
print(item['hostname'])
|
||||
print(item['filename'])
|
||||
return json_output
|
||||
#endpoint = url + '/v1/logging/exechistories'
|
||||
#payload_dict = {
|
||||
# "type":[1, 2, 6, 7],
|
||||
# "checkpoint":"000000000000000000000000",
|
||||
# "policy": [policiesnames[choice]]
|
||||
#}
|
||||
#payload = json.dumps(payload_dict)
|
||||
#print(payload)
|
||||
#response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
|
||||
#parse_text = json.loads(response.text)
|
||||
#text_response = checkpoint_stomper(parse_text['response']['exechistories'], url, policiesnames[choice])
|
||||
#
|
||||
#if outputjson == False:
|
||||
# return response
|
||||
#
|
||||
#parse_text = json.loads(response.text)
|
||||
#
|
||||
#for item in parse_text['response']['exechistories']:
|
||||
# print(item['checkpoint'])
|
||||
# print(item['datetime'])
|
||||
# print(item['hostname'])
|
||||
# print(item['filename'])
|
||||
# checkpoint_stomper(item['checkpoint'], endpoint, headers, policiesnames[choice])
|
||||
|
||||
def checkpoint_stomper(checkpoint, endpoint, headers, policyname):
|
||||
print(checkpoint)
|
||||
def checkpoint_stomper(checkpoint, url, policy, headers):
|
||||
endpoint = url + '/v1/logging/exechistories'
|
||||
payload_dict = {
|
||||
"type":[1,2,6,7],
|
||||
"checkpoint": checkpoint,
|
||||
"policy":[policyname]
|
||||
"policy": [policy]
|
||||
}
|
||||
payload = json.dumps(payload_dict)
|
||||
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
|
||||
parse_text = json.loads(response.text)
|
||||
# Send Whole JSON Response
|
||||
for item in parse_text['response']['exechistories']:
|
||||
print("test")
|
||||
return parse_text
|
||||
#for index, item in enumerate(parse_text):
|
||||
# if index == len(parse_text) - 1:
|
||||
# checkpoint = item['checkpoint']
|
||||
# print(f"Time: {item['datetime']} Checkpoint: {item['checkpoint']}")
|
||||
# response_fuzzer(checkpoint, url, policyname)
|
||||
# else:
|
||||
# if (datetime.date.today() - datetime.timedelta(days=30) > datetime.datetime.strptime(item['datetime'].replace( ' +0000 UTC', ''), '%Y-%m-%dT%H:%M:%SZ').date()):
|
||||
# pass
|
||||
# else:
|
||||
# response_fuzzer(checkpoint, url, policyname)
|
||||
|
||||
|
||||
print("Finished")
|
||||
+24
-14
@@ -8,12 +8,12 @@ def aggregateHashes(executions_json) -> pd.DataFrame:
|
||||
"""
|
||||
Takes the executions, aggregates all the data with sha256 as primary, then returns aggregated dataframe
|
||||
"""
|
||||
data = executions_json.json()
|
||||
data = json.loads(executions_json)
|
||||
df = pd.DataFrame(data["response"]["exechistories"])
|
||||
|
||||
if df.empty:
|
||||
return df
|
||||
|
||||
print(df)
|
||||
# Aggregate by sha256, deduplicate lists, and preserve order
|
||||
agg_df = df.groupby("sha256").agg(lambda x: list(dict.fromkeys(x))).reset_index()
|
||||
|
||||
@@ -64,16 +64,17 @@ def augmentAggregatedHashes(url, agg_df: pd.DataFrame) -> pd.DataFrame:
|
||||
|
||||
df_api = pd.DataFrame(rows)
|
||||
|
||||
aug_df = agg_df.merge(df_api, on="sha256", how="left")
|
||||
|
||||
df = agg_df.merge(df_api, on="sha256", how="left")
|
||||
aug_df = df[['sha256', 'filename_x', 'description', 'productname', 'productversion', 'publisher_y', 'publisher_x', 'netdomain', 'hostname', 'username', 'pprocess', 'gprocess', 'commandline', 'reputation_lastseen', 'reputation_scannercount', 'reputation_scannermatch', 'reputation_status', 'reputation_threatlevel', 'reputation_threatname', 'reputation_timestamp']]
|
||||
return aug_df
|
||||
|
||||
def categorizeHashes(aug_df: pd.DataFrame, threat_tolerance: int, untrusted_publishers: list):
|
||||
if untrusted_publishers is None:
|
||||
untrusted_publishers = []
|
||||
if untrusted_publishers is None:
|
||||
untrusted_publishers = []
|
||||
|
||||
df = aug_df.copy()
|
||||
def reputationtool(row, threat_tolerance):
|
||||
df = aug_df.copy()
|
||||
|
||||
def reputationtool(row, threat_tolerance):
|
||||
if row["reputation_scannermatch"] == "N/A":
|
||||
return True
|
||||
try:
|
||||
@@ -83,11 +84,20 @@ def categorizeHashes(aug_df: pd.DataFrame, threat_tolerance: int, untrusted_publ
|
||||
pass
|
||||
return False
|
||||
|
||||
mask_needsreview = (df["publisher_y"] == "Not Signed") & df.apply(lambda row: reputationtool(row, threat_tolerance), axis=1)
|
||||
mask_approved = (df["publisher_y"] != "Not Signed") & (~df["publisher_y"].isin(untrusted_publishers))
|
||||
mask_needsreview = (df["publisher_y"] == "Not Signed") & df.apply(lambda row: reputationtool(row, threat_tolerance), axis=1)
|
||||
|
||||
needsreview_df = df[mask_needsreview]
|
||||
approved_df = df[mask_approved]
|
||||
remaining_df = df[~(mask_needsreview | mask_approved)]
|
||||
mask_approved = (
|
||||
((df["publisher_y"] != "Not Signed") & (~df["publisher_y"].isin(untrusted_publishers))) &
|
||||
(df["publisher_y"] != "Not Signed") # explicitly signed
|
||||
) | (
|
||||
(df["publisher_y"] == "Not Signed") &
|
||||
(~df.apply(lambda row: reputationtool(row, threat_tolerance), axis=1)) &
|
||||
(~df["publisher_y"].isin(untrusted_publishers)) # exclude untrusted even if unsigned
|
||||
)
|
||||
|
||||
return needsreview_df, approved_df, remaining_df
|
||||
|
||||
needsreview_df = df[mask_needsreview]
|
||||
approved_df = df[mask_approved]
|
||||
remaining_df = df[~(mask_needsreview | mask_approved)]
|
||||
|
||||
return needsreview_df, approved_df, remaining_df
|
||||
|
||||
+27
-1
@@ -41,18 +41,44 @@ def filepathInitialGroup(df: pd.DataFrame):
|
||||
return join_parts(prefix)
|
||||
|
||||
# Step 6: Group directories by shared prefix using custom logic
|
||||
"""
|
||||
Loop through each directory path
|
||||
directories: list of all directory paths.
|
||||
groups: will hold lists of grouped directories.
|
||||
used: tracks which directories have already been grouped.
|
||||
"""
|
||||
directories = df["directory"].tolist()
|
||||
groups = []
|
||||
used = set()
|
||||
|
||||
#For Each directory, compare it with others
|
||||
"""
|
||||
Skip if already grouped.
|
||||
Start a new group with the current path.
|
||||
parts_i is the list of folder names in the path (e.g., ["C:", "Users", "John", "Documents"]).
|
||||
"""
|
||||
|
||||
for i, path in enumerate(directories):
|
||||
if path in used:
|
||||
continue
|
||||
group = [path]
|
||||
parts_i = get_parts(path)
|
||||
|
||||
#Compare with all other directories: For each other directory, split it into parts and find the common prefix (shared folder structure).
|
||||
"""
|
||||
Logic:
|
||||
If the directory is deep (>3 parts) and shares at least 3 parts → group it.
|
||||
If it's exactly 3 parts long and shares at least 2 → group it.
|
||||
Or, if it shares all but one part and is deep → group it.
|
||||
These rules are designed to:
|
||||
Group directories that are closely related in structure.
|
||||
Avoid grouping unrelated paths that just happen to start similarly.
|
||||
"""
|
||||
|
||||
for j in range(i + 1, len(directories)):
|
||||
parts_j = get_parts(directories[j])
|
||||
common = os.path.commonprefix([parts_i, parts_j])
|
||||
#Apply grouping rules
|
||||
if (len(parts_i) > 3 and len(common) >= 3) or (len(parts_i) == 3 and len(common) >= 2):
|
||||
group.append(directories[j])
|
||||
used.add(directories[j])
|
||||
@@ -81,7 +107,7 @@ def filepathInitialGroup(df: pd.DataFrame):
|
||||
path_ineligible = grouped_df[grouped_df["depth"] <= 2].drop(columns=["depth"])
|
||||
|
||||
# Step 10: Move entries from eligible to ineligible if grouped_directory contains 'C:\Users' or 'c$\Users'
|
||||
mask = path_eligible["grouped_directory"].str.contains(r"(?i)(?:\\Users|\\c\$\\Users)")
|
||||
mask = path_eligible["grouped_directory"].str.contains(r"(?i)(?:\\Users|\\c\$\\Users|inetpub\\wwwroot|windows\\temp)", na=False)
|
||||
move_to_ineligible = path_eligible[mask]
|
||||
path_eligible = path_eligible[~mask]
|
||||
path_ineligible = pd.concat([path_ineligible, move_to_ineligible], ignore_index=True)
|
||||
|
||||
Reference in New Issue
Block a user