diff --git a/.gitignore b/.gitignore index 7e0cdf3..848b851 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,7 @@ .env *.html +<<<<<<< HEAD *.csv +======= +>>>>>>> ccbf716 (Zar-Branch (#6)) *__pycache__* \ No newline at end of file diff --git a/AirlockTools.py b/AirlockTools.py index 3684d37..8e400ab 100644 --- a/AirlockTools.py +++ b/AirlockTools.py @@ -20,7 +20,10 @@ import utils.allowlist import utils.hashfunctions import utils.pathfunctions import utils.allowfunctions +<<<<<<< HEAD import utils.colortext as ct +======= +>>>>>>> ccbf716 (Zar-Branch (#6)) import urllib3 import pandas as pd @@ -77,6 +80,7 @@ def menu_main(): else: print(ct.colorText("Invalid choice. Please try again.","red")) +<<<<<<< HEAD def menu_local_approve(): while True: print("\n--- Submenu ---") @@ -158,6 +162,59 @@ def menu_prepare_to_enforce(): print(ct.colorText(" [✗] This step has not been completed","red")) print(ct.colorText("4. Add hash threat information to list of executions", "cyan")) +======= + augmented.to_html("z_augmented_list.html", index=False) + + badpublisherlist = [] + categorized = utils.hashfunctions.categorizeHashes(augmented, 5, badpublisherlist) + categorized[0].to_html("z_needs_review.html", index=False) + categorized[1].to_html("z_approved_hashes.html", index=False) + categorized[2].to_html("z_remaining.html", index=False) + + if choice == '4': + html_file = "z_augmented_list.html" + augmented_df = pd.read_html(html_file) + print(augmented_df) + combined_df = pd.concat(augmented_df, ignore_index=True) + + path_eligible, path_ineligible = utils.pathfunctions.filepathInitialGroup(combined_df) + path_eligible.to_html("z_eligble_paths.html", index=False) + path_ineligible.to_html("z_ineligible_paths.html",index=False) + + if choice == '5': + + executionhist = utils.allowlist.allowlistexechistories(url,True) + #print(executionhist) + + aggregated = utils.hashfunctions.aggregateHashes(executionhist) + print(aggregated) + + augmented = utils.hashfunctions.augmentAggregatedHashes(url,aggregated) + print(augmented) + + augmented.to_html("z_augmented_list.html", index=False) + html_file = "z_augmented_list.html" + augmented_df = pd.read_html(html_file) + combined_df = pd.concat(augmented_df, ignore_index=True) + + path_eligible, path_ineligible = utils.pathfunctions.filepathInitialGroup(combined_df) + path_eligible.to_html("z_eligble_paths.html", index=False) + path_ineligible.to_html("z_ineligible_paths.html",index=False) + + badpublisherlist = ["Brave Software, Inc.", "Zoom Video Communications, Inc."] + categorized = utils.hashfunctions.categorizeHashes(augmented, 5, badpublisherlist) + categorized[0].to_html("z_needs_review.html", index=False) + categorized[1].to_html("z_approved_hashes.html", index=False) + categorized[2].to_html("z_remaining.html", index=False) + + allowpaths = utils.allowfunctions.filter_and_drop(categorized[1],path_eligible,4) + allowpaths.to_html("z_allowed_paths.html", index=False) + + + + + +>>>>>>> ccbf716 (Zar-Branch (#6)) if os.path.exists(f"dataframe_csv\\df_augmented_combo_{first_policy}_{second_policy}.csv") == True: print(ct.colorText(" [✓] This step has been completed","green")) diff --git a/hashtest.py b/hashtest.py new file mode 100644 index 0000000..b0dab88 --- /dev/null +++ b/hashtest.py @@ -0,0 +1,5 @@ +hashes = '' +while True: + inputhash = input("Hash: ") + hashes = hashes + ',' + inputhash + print(hashes) \ No newline at end of file diff --git a/utils/allowlist.py b/utils/allowlist.py index 6cc6e2d..80427ef 100644 --- a/utils/allowlist.py +++ b/utils/allowlist.py @@ -17,7 +17,10 @@ import requests import json import os import time +<<<<<<< HEAD import utils.colortext as ct +======= +>>>>>>> ccbf716 (Zar-Branch (#6)) def pullPolicyExechistories(url, choice, policiesnames, outputjson: bool): @@ -73,4 +76,75 @@ def listPolicies(url): policyids.append(list['groupid']) choice = input(ct.colorText("Select Policy Group: ", "white")) choice = int(choice) - 1 - return choice, policiesnames \ No newline at end of file +<<<<<<< HEAD + return choice, policiesnames +======= + 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 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, url, policy, headers): + endpoint = url + '/v1/logging/exechistories' + payload_dict = { + "type":[1,2,6,7], + "checkpoint": checkpoint, + "policy": [policy] + } + payload = json.dumps(payload_dict) + response = requests.request("POST", endpoint, headers=headers, data=payload, verify=False) + parse_text = json.loads(response.text) + 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") +>>>>>>> ccbf716 (Zar-Branch (#6)) diff --git a/utils/hashfunctions.py b/utils/hashfunctions.py index c193284..43481b8 100644 --- a/utils/hashfunctions.py +++ b/utils/hashfunctions.py @@ -90,15 +90,22 @@ def categorizeHashes(aug_df: pd.DataFrame, threat_tolerance: int, untrusted_publ df = aug_df.copy() +<<<<<<< HEAD def reputationtool(row): val = row["reputation_scannermatch"] if pd.isna(val) or val == "N/A": return row["publisher_y"] == "Not Signed" +======= + def reputationtool(row, threat_tolerance): + if row["reputation_scannermatch"] == "N/A": + return True +>>>>>>> ccbf716 (Zar-Branch (#6)) try: return int(val) > threat_tolerance except (ValueError, TypeError): return row["publisher_y"] == "Not Signed" +<<<<<<< HEAD df["reputation_flag"] = df.apply(reputationtool, axis=1) mask_needsreview = ( @@ -125,3 +132,22 @@ def categorizeHashes(aug_df: pd.DataFrame, threat_tolerance: int, untrusted_publ unapproved_df = df[~(mask_needsreview | mask_approved)] return needsreview_df, approved_df, unapproved_df +======= + 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))) & + (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 + ) + + + needsreview_df = df[mask_needsreview] + approved_df = df[mask_approved] + remaining_df = df[~(mask_needsreview | mask_approved)] + + return needsreview_df, approved_df, remaining_df +>>>>>>> ccbf716 (Zar-Branch (#6)) diff --git a/utils/pathfunctions.py b/utils/pathfunctions.py index ab8e166..890f62f 100644 --- a/utils/pathfunctions.py +++ b/utils/pathfunctions.py @@ -57,21 +57,56 @@ def filepathInitialGroup(df: pd.DataFrame): break return join_parts(prefix) +<<<<<<< HEAD # Step 6: Group directories by shared 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. + """ +>>>>>>> ccbf716 (Zar-Branch (#6)) 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) +<<<<<<< HEAD for j in range(i + 1, len(directories)): parts_j = get_parts(directories[j]) common = os.path.commonprefix([parts_i, parts_j]) +======= + #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 +>>>>>>> ccbf716 (Zar-Branch (#6)) 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]) @@ -99,7 +134,11 @@ def filepathInitialGroup(df: pd.DataFrame): path_eligible = grouped_df[grouped_df["depth"] > 2].drop(columns=["depth"]) path_ineligible = grouped_df[grouped_df["depth"] <= 2].drop(columns=["depth"]) +<<<<<<< HEAD # Step 10: Move entries from eligible to ineligible if grouped_directory contains excluded directories +======= + # Step 10: Move entries from eligible to ineligible if grouped_directory contains 'C:\Users' or 'c$\Users' +>>>>>>> ccbf716 (Zar-Branch (#6)) 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]