mysticmomba/AirlockTools
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 3 Wiki Activity

Major refactor: security enhancements, modularization, config integration, reduced Parquet reliance

Browse Source 
- Migrated codebase to class-based architecture for better modularity and maintainability
- Introduced system_config.json for centralized configuration (required for runtime)
- Added structured working directories for improved file organization
- Significantly reduced reliance on Parquet; replaced with alternative data handling
- Implemented security improvements across modules
- Several TODOs remain in the main script for future enhancements
- Linter formatting affected readability in some files (e.g., utils); cleanup is on the agenda
This commit is contained in:
Zarithas
2025-10-05 22:20:42 -04:00
parent b1e6af01c6
commit 89db386ffe
26 changed files with 3530 additions and 2389 deletions
Download Patch File Download Diff File Expand all files Collapse all files
utils/policyfunctions.py
-657
View File
@@ -1,657 +0,0 @@
# Copyright (C) 2025 James Brotosky, Brandon Wickline
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published
# by the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
#Local Imports
import utils.utils as ct
import utils.hashfunctions as hashf
import utils.pathfunctions as pathf
import utils.policyfunctions as policyf
#Standard Libary Imports:
import datetime
import gc
import json
import os
import re
import sys
#3rd Party Imports:
import ijson
import pandas as pd
import requests
import tqdm
from bson import ObjectId
def addHash(url, policy, hash):
print(f"Adding the following: {hash} \n to {policy}:")
for p in hash:
pass
# print(p)
def addPath(url, policy, hash):
print(f"Adding the following Path Exclusions to {policy}:")
for p in hash:
print(p)
def addPub(url, policy, publist):
print(f"Adding the following Publishers to {policy}:")
for p in publist:
print(p)
def addHashReal(url, allowlistID, hashlist):
endpoint = url + '/v1/hash/application/add'
payload = {
"applicationid" : allowlistID,
"hashes" : hashlist
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
payload = json.dumps(payload)
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
response.raise_for_status() # Raise an error for bad status codes
parse_text = json.loads(response.text)
print(parse_text)
def addPathReal(url, grouplistID, pathlist):
endpoint = url + '/v1/group/path/add'
payload = {
"groupid" : grouplistID,
"path" : pathlist
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
print(payload)
payload = json.dumps(payload)
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
print(response.text)
def addPubReal(url, grouplistID, publist):
endpoint = url + '/v1/group/publisher/add'
payload = {
"groupid" : grouplistID,
"publisher" : publist
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
print(payload)
payload = json.dumps(payload)
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
print(response.text)
def getPolicyInfo(url, policy, type, days, parquet=True):
executionhist_policy = pd.DataFrame()
exehist = pullPolicyExechistories(url, policy, type, days, True)
if exehist is not None:
data = json.loads(exehist)
executionhist_policy = pd.DataFrame(data["response"]["exechistories"])
if not executionhist_policy.empty:
executionhist_policy = executionhist_policy[['datetime','sha256', 'publisher', 'filename', 'hostname', 'username', 'pprocess', 'gprocess', 'commandline']]
executionhist_policy['policy'] = policy # Add policy column here
executionhist_policy = executionhist_policy.drop_duplicates(subset=['sha256', 'filename', 'hostname'])
executionhist_policy = executionhist_policy.sort_values(by=['sha256', 'filename'])
if parquet:
executionhist_policy.to_parquet(f"prepare_policy\\parquet\\execution_history_{policy}.parquet", index=False)
print(ct.colorText(f"Staging of Execution history for policy: {policy} is complete", "green"))
del data
del exehist
gc.collect()
return executionhist_policy
def sendToPolicy(url, paths, hashes, publishers, destination_name, destination_id, allowlist_name, allowlist_id):
pathexclusions = pd.read_parquet(paths)
allowbyhash = pd.read_parquet(hashes)
publishers = ct.tryToReadCSV(publishers)
ct.areYouSure()
confirmation = input(ct.colorText("Type 'I AGREE' to continue: ","white"))
if confirmation.strip().upper() == "I AGREE":
print(ct.colorText("Proceeding with the code...", "yellow"))
print(ct.colorText(f"Adding path exclusions to {destination_name}", "yellow"))
# Get unique combinations of longestcfp and file_extension
unique_combinations = pathexclusions[["longestcfp", "file_extension"]].drop_duplicates()
# Regex to match a Windows drive letter at the start (e.g., C:\)
drive_letter_pattern = re.compile(r'^[a-zA-Z]:\\')
# Build processed paths like \\path\\**.exe or C:\path\**.jar
processed_paths = [
(path if drive_letter_pattern.match(path) else f"\\\\{path}") + f"\\**{ext}"
for path, ext in unique_combinations.itertuples(index=False, name=None)
]
addPathReal(url, destination_id,processed_paths)
print(ct.colorText(f"Adding publishers to {destination_name}", "yellow"))
if publishers.empty:
print(ct.colorText("The publishers list is empty.", "red"))
else:
publisher_list = publishers['publisher'].tolist()
addPubReal(url, destination_id, publisher_list)
print(ct.colorText(f"These hashes would be added to {allowlist_name}", "yellow"))
allowlist = allowbyhash['sha256'].unique().tolist()
addHash(url, allowlist_id,allowlist)
ct.locked()
exit()
else:
print(ct.colorText("Operation aborted. You MUST EXPLICITLY AGREE to proceed.", "red"))
def pullPolicyExechistories(url, policiesnames, type, days, outputjson: bool):
file_path = 'chunkinator.json'
if not os.path.exists(file_path):
with open(file_path, 'w') as file:
json.dump({'error': 'Success', 'response': {'exechistories': []}}, file)
print(f"File '{file_path}' has been created.")
else:
print(f"File '{file_path}' already exists.")
headers = {"X-APIKey": os.getenv('APIKEY')}
checkpoint = str(skipback(days))
json_output = {'error': 'Success', 'response': {'exechistories': []}}
with tqdm.tqdm(file=sys.stdout, leave=True, total=10000, desc=f"Checkpoint Progess: {checkpoint}", colour="blue", initial=1) as filebar:
with tqdm.tqdm(file=sys.stdout, leave=True, total=100, desc=f"Total of {policiesnames} Complete: ") as pbar:
while True:
item = {}
json_response_data = checkpoint_stomper(checkpoint, url, type, policiesnames, headers)
histories = json_response_data['response']['exechistories']
filebar.total=len(histories)
if not histories:
break
match_found = True
if match_found == True:
for index, item in enumerate(histories):
if index == len(histories) - 1:
checkpoint = item['checkpoint']
filebar.desc = f"Checkpoint Progress: {checkpoint}"
break
else:
if (datetime.date.today() - datetime.timedelta(days=days) > datetime.datetime.strptime(item['datetime'].replace(' +0000 UTC', ''), '%Y-%m-%dT%H:%M:%SZ').date()):
pass
else: json_output['response']['exechistories'].append(item)
filebar.update(1)
filebar.refresh()
seen = {}
if os.path.exists(file_path):
with open(file_path, 'r') as file:
existing_data = json.load(file)
combined = existing_data['response']['exechistories'] + json_output['response']['exechistories']
else:
combined = json_output['response']['exechistories']
for item in combined:
key = (item.get('sha256'), item.get('filename'), item.get('hostname'))
seen[key] = item
deduplicated = list(seen.values())
with open(file_path, 'w') as file:
json.dump({'error': 'Success', 'response': {'exechistories': deduplicated}}, file)
json_output['response']['exechistories'].clear()
date_diff = datetime.date.today() - datetime.datetime.strptime(item['datetime'].replace(' +0000 UTC', ''), '%Y-%m-%dT%H:%M:%SZ').date()
percentage_diff = (((days + 10) - date_diff.days) / (days + 10)) * 100
pbar.n = round(percentage_diff)
pbar.set_description_str(f"Total of {policiesnames} Complete: ")
pbar.refresh()
filebar.n = 1
with open(file_path, 'r') as file:
final_output = json.load(file)
os.remove(file_path)
return json.dumps(final_output) if outputjson else None
def checkpoint_stomper(checkpoint, url, type, policy, headers):
json_output = {'error': 'Success', 'response': {'exechistories': []}}
endpoint = url + '/v1/logging/exechistories'
payload_dict = {
"type":[type],
"checkpoint": checkpoint,
"policy": [policy]
}
payload = json.dumps(payload_dict)
with requests.request("POST", endpoint, headers=headers, data=payload, verify=False, stream=True) as response:
parser = ijson.items(response.raw, 'response.exechistories.item')
for item in parser:
key = (item.get('sha256'), item.get('hostname'))
if key not in json_output:
json_output['response']['exechistories'].append(item)
parse_text = json.loads(json.dumps(json_output))
return parse_text
def listPolicies(url):
endpoint = url + '/v1/group'
print(ct.colorText("[+] Grabbing All Policies", "cyan"))
payload = {}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
parse_text = json.loads(response.text)
return parse_text
def choosePolicies(url):
allpolicies = listPolicies(url)
policiesnames = []
policyids = []
for index, list in enumerate(allpolicies['response']['groups'], start=1):
print(ct.colorText(f"{index}. {list['name']}", "yellow"))
policiesnames.append(list['name'])
policyids.append(list['groupid'])
choice = input(ct.colorText("Select Policy Group: ", "white"))
choice = int(choice) - 1
return choice, policiesnames, policyids
def getPolicyDataframe(url) -> pd.DataFrame:
return pd.DataFrame(listPolicies(url)['response']['groups'])
def listATPolicies(url):
endpoint = url + '/v1/group'
print(ct.colorText("[+] Grabbing All Policies", "cyan"))
headers = {
"X-APIKey": os.getenv('APIKEY')
}
try:
response = requests.post(endpoint, headers=headers, json={}, verify=False)
response.raise_for_status()
parse_text = response.json()
at_policies = {}
for index, group in enumerate(parse_text.get('response', {}).get('groups', []), start=1):
name = group.get('name', '')
if "AT" in name:
print(ct.colorText(f"{index}. {name}", "yellow"))
at_policies[name] = group.get('groupid')
return at_policies
except requests.exceptions.RequestException as e:
print(ct.colorText(f"[!] Request failed: {e}", "red"))
return {}
except (KeyError, json.JSONDecodeError) as e:
print(ct.colorText(f"[!] Failed to parse response: {e}", "red"))
return {}
def listAllowlists(url: str) -> tuple[int, list, list]:
endpoint = url + '/v1/application'
print(ct.colorText("[+] Grabbing All Allowlists", "cyan"))
payload = {}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
parse_text = json.loads(response.text)
policiesnames = []
policyids = []
for index, item in enumerate(parse_text['response']['applications'], start=1):
if index >= 38:
print(ct.colorText(f"{index}. {item['name']}", "yellow"))
policiesnames.append(item['name'])
policyids.append(item['applicationid'])
while True:
try:
choice = int(input(ct.colorText("Select allowlist: ", "white")))
if choice < 38 or choice > len(parse_text['response']['applications']):
print(ct.colorText("Please only choose an allowlist designed for this use - '38+'", "red"))
else:
adjusted_choice = choice - 38
return adjusted_choice, policiesnames, policyids
except ValueError:
print(ct.colorText("Invalid input. Please enter a number.", "red"))
def skipback(days):
"""
Generate a MongoDB ObjectId for a given number of days ago from today.
"""
adjusted_days = days
date_days_ago = datetime.datetime.now(datetime.UTC) - datetime.timedelta(days=adjusted_days)
timestamp = int(date_days_ago.timestamp())
hex_timestamp = format(timestamp, '08x')
objectid_hex = hex_timestamp + '0000000000000000'
return ObjectId(objectid_hex)
def sendToPolicyTest(url, paths, hashes, publishers, destination_name, destination_id, allowlist_id, allowlist_name):
pathexclusions = pd.read_parquet(paths)
allowbyhash = pd.read_parquet(hashes)
publishers = ct.tryToReadCSV(publishers)
print(ct.colorText(f"These path exclusions would be added to {destination_name}", "yellow"))
# Get unique combinations of longestcfp and file_extension
unique_combinations = pathexclusions[["longestcfp", "file_extension"]].drop_duplicates()
# Regex to match a Windows drive letter at the start (e.g., C:\)
drive_letter_pattern = re.compile(r'^[a-zA-Z]:\\')
# Build processed paths like \\path\\**.exe or C:\path\**.jar
processed_paths = [
(path if drive_letter_pattern.match(path) else f"\\\\{path}") + f"\\**{ext}"
for path, ext in unique_combinations.itertuples(index=False, name=None)
]
addPath(url, destination_id,processed_paths)
print(ct.colorText(f"These publishers would added to {destination_name}", "yellow"))
if publishers.empty:
print(ct.colorText("The publishers list is empty.", "red"))
else:
publisher_list = publishers['publisher'].tolist()
addPub(url, destination_id, publisher_list)
print(ct.colorText(f"These hashes would be added to {allowlist_name}", "yellow"))
allowlist = allowbyhash['sha256'].unique().tolist()
addHash(url, allowlist_id,allowlist)
def updateAuditPoliciesFromEnforcementPolices(url, policy_relationship_map):
for enforcement_policy, audit_policy in policy_relationship_map.items():
assignPoliciesfromGroup(url, enforcement_policy, audit_policy)
turnOnAudit(url, audit_policy)
def assignPoliciesfromGroup(url, source_policy_id, target_policy_id):
endpoint = url + '/v1/group/assign'
payload = {
"groupid" : {source_policy_id},
"targetgroupid" : {target_policy_id}
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
parse_text = json.loads(response.text)
print(parse_text)
def turnOnAudit(url, policyid):
endpoint = url + '/v1/group/settings/auditmode'
payload = {
"groupid" : {policyid},
"auditmode" : "1"
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
parse_text = json.loads(response.text)
print(parse_text)
def agentsInPolicy(url, policyid):
endpoint = url + '/v1/group/agents'
payload = {
"groupid" : {policyid}
}
headers = {
"X-APIKey": os.getenv('APIKEY')
}
response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False)
parse_text = json.loads(response.text)
print(parse_text)
def prepare_to_enforce(url, bad_publisher_list, pups, badpathparts, threat_tolerance_constant = 4, path_exclusion_constant = 4, min_files_for_path = 4):
destination_name = " "
destination_id = " "
allowlist_name = " "
allowlist_id = " "
policylist = []
type = [1, 2, 6, 7]
parq_base_dir = "prepare_policy\\parquet\\"
appr_base_dir = "prepare_policy\\approved\\"
needappr_base_dir = "prepare_policy\\needs_approved\\"
pflight_base_dir = "prepare_policy\\preflight\\"
#If the directorys where we're going to store our output dont exist, make them.
os.makedirs(parq_base_dir, exist_ok=True)
os.makedirs(needappr_base_dir, exist_ok=True)
os.makedirs(appr_base_dir, exist_ok=True)
os.makedirs(pflight_base_dir, exist_ok=True)
while True:
ct.printEnforceChecklist(parq_base_dir,appr_base_dir, needappr_base_dir, pflight_base_dir, policylist, allowlist_name, destination_name)
choice = input(ct.colorText("\nEnter your choice: ", "white"))
if choice == "1":
policylist, policyids = getMultiplePolicySelections(url)
elif choice == "2":
print(ct.colorText(f"Please choose destination_name Policy for Path Exclusions","white"))
choice, policynames, policyid = choosePolicies(url)
#print(allowlist_parent_tuple)
destination_name = policynames[choice]
destination_id = policyid[choice]
print(ct.colorText(f"Please choose Allowlist for Hashes","white"))
choice, allowlists,allowid = listAllowlists(url)
#print(allowlist_parent_tuple)
allowlist_name = allowlists[choice]
allowlist_id = allowid[choice]
print(destination_name, allowlist_name)
elif choice == "3":
policyf.buildExecHistory(url,
policylist,
parq_base_dir,
needappr_base_dir,
type,
threat_tolerance_constant,
bad_publisher_list,
pups,
)
csvs = [f"{needappr_base_dir}unknown_hashes.csv", f"{needappr_base_dir}good_hashes.csv"]
#Since we want to build paths as if they were all in the same policy to begin with, lets group them that way
for csv in csvs:
df = ct.tryToReadCSV(csv)
df['policy'] = destination_name
df.to_csv(csv)
elif choice == "4":
pathf.generateApprovables(needappr_base_dir, appr_base_dir, parq_base_dir, badpathparts, bad_publisher_list, path_exclusion_constant, min_files_for_path)
elif choice == "5":
savePreFlights(appr_base_dir, parq_base_dir, pflight_base_dir)
elif choice == "6":
if os.path.exists(f"{pflight_base_dir}final_path_exclusions.html") and os.path.exists(f"{pflight_base_dir}\\final_hash_approvals.html") and allowlist_name != " " and destination_name != " ":
sendToPolicyTest(
url,
f"{parq_base_dir}final_path_exclusions.parquet",
f"{parq_base_dir}final_hash_approvals.parquet",
f"{appr_base_dir}publishers.parquet",
destination_name,
destination_id,
allowlist_name,
allowlist_id
)
elif choice == "7":
if os.path.exists(f"{pflight_base_dir}final_path_exclusions.html") and os.path.exists(f"{pflight_base_dir}\\final_hash_approvals.html") and allowlist_name != " " and destination_name != " ":
sendToPolicy(
url,
f"{parq_base_dir}final_path_exclusions.parquet",
f"{parq_base_dir}final_hash_approvals.parquet",
f"{appr_base_dir}publishers.parquet",
destination_name,
destination_id,
allowlist_name,
allowlist_id,
)
elif choice == "R":
pathf.clean_folders(parq_base_dir, appr_base_dir, needappr_base_dir, pflight_base_dir)
elif choice == "Q":
break
else:
print(ct.colorText("Invalid choice. Please try again.", "red"))
def buildExecHistory(url,
policylist,
parq_base_dir,
needappr_base_dir,
type,
threat_tolerance_constant,
bad_publisher_list,
pups
):
exe_hist_parq_list = []
while True:
try:
history_days = int(input("Enter the how many days in of history do you want to pull - select a number between 1 and 150: "))
if 1 <= history_days <= 150:
break
else:
print("Invalid input. Please enter a number between 1 and 150.")
except ValueError:
print("Invalid input. Please enter a valid integer.")
for policy in policylist:
policy_exec_history = policyf.getPolicyInfo(url, policy, type, history_days)
policy_exec_history['policy'] = policy
policy_exec_history.to_parquet(f"{parq_base_dir}Exec_Hist_{policy}.parquet")
exe_hist_parq_list.append(f"{parq_base_dir}Exec_Hist_{policy}.parquet")
augmented_hashlist = hashf.combineHashes(url, exe_hist_parq_list)
augmented_hashlist.to_parquet(f"{parq_base_dir}augmentedHashlist.parquet",index=False)
needsreview_df, approved_df, unapproved_df = hashf.categorizeHashes(augmented_hashlist, threat_tolerance_constant, bad_publisher_list, pups)
needsreview_df.to_parquet(f"{parq_base_dir}needsreview.parquet",index=False)
approved_df.to_parquet(f"{parq_base_dir}approved.parquet",index=False)
unapproved_df.to_parquet(f"{parq_base_dir}unapproved.parquet",index=False)
condensed_executions = hashf.condenseExecutions(exe_hist_parq_list)
condensed_executions.to_parquet(f"{parq_base_dir}condensed_executions.parquet", index=False)
unknown, good, bad = hashf.divideSortedHashExecutions(
f"{parq_base_dir}needsreview.parquet",
f"{parq_base_dir}approved.parquet",
f"{parq_base_dir}unapproved.parquet",
f"{parq_base_dir}condensed_executions.parquet",
pups
)
dataframes = {
"unknown_hashes" : unknown,
"good_hashes": good,
"bad_hashes": bad
}
for name, df in dataframes.items():
df.to_csv(f"{needappr_base_dir}{name}.csv", index=False)
df.to_parquet(f"{parq_base_dir}{name}.parquet", index=False)
ct.style_dataframe_dark(df, f"{needappr_base_dir}{name}.html")
def savePreFlights(appr_base_dir, parq_base_dir, pflight_base_dir):
if os.path.exists(f"{appr_base_dir}primary_Paths.csv"):
if not os.path.exists(f"{parq_base_dir}final_hash_approvals.parquet") and not os.path.exists(f"{parq_base_dir}final_path_exclusions.parquet"):
pathexclusions, allowbyhash = hashf.generatePreflights(
f"{parq_base_dir}all_hashes.parquet",
f"{appr_base_dir}primary_Paths.csv",
f"{appr_base_dir}secondary_Paths.csv")
dataframes = {
"final_path_exclusions" : pathexclusions,
"final_hash_approvals": allowbyhash
}
for name, df in dataframes.items():
df.to_csv(f"{pflight_base_dir}{name}.csv", index=False)
df.to_parquet(f"{parq_base_dir}{name}.parquet", index=False)
ct.style_dataframe_dark(df, f"{pflight_base_dir}{name}.html")
def getMultiplePolicySelections(url):
policynameslist = []
policyidlist = []
while True:
choice, policynames, policyid = listPolicies(url)
selected_policy = policynames[choice]
selected_policyid = policyid[choice]
if selected_policy not in policynameslist:
policynameslist.append(selected_policy)
if selected_policyid not in policyidlist:
policyidlist.append(selected_policyid)
while True:
answer = input(ct.colorText("Do you want to load another policy? (yes/no): ", "white")).strip().lower()
if answer in ("no", "n"):
break # Exit the inner loop and then the outer loop
elif answer in ("yes", "y"):
break # Exit the inner loop and continue the outer loop
else:
print(ct.colorText("Please answer with 'yes' or 'no'.", "red"))
if answer in ("no", "n"):
break
return policynameslist, policyidlist
def getDestAllowlist(url, groupid):
allowlists = getPolicyAllowlists(url, groupid)
matches = allowlists[
allowlists['name'].str.contains('local', case=False, na=False) &
allowlists['name'].str.contains('approval', case=False, na=False),
'applicationid'
].values
app_id = matches[0] if len(matches) > 0 else None
return app_id
def getPolicyAllowlists(url, groupid):
endpoint = url + '/v1/group/policies'
payload = {
"groupid" : f"{groupid}"
}
headers = {"X-APIKey": os.getenv('APIKEY')}
payload = json.dumps(payload)
response = requests.post(endpoint, headers=headers, data=payload, verify=False)
result = json.loads(response.text)
data = pd.DataFrame(result["response"]["applications"])
return data