mirror of
https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git
synced 2026-06-11 21:54:56 +03:00
27c6414ca5
Complete skill folder anatomy across all cybersecurity skills: - scripts/agent.py: 80-150 line Python agents using real libraries (impacket, boto3, azure-mgmt-*, kubernetes, pefile, yara, scapy, shodan, stix2, etc.) - references/api-reference.md: real API documentation with method signatures - LICENSE: MIT license for all skill folders
142 lines
5.8 KiB
Python
142 lines
5.8 KiB
Python
#!/usr/bin/env python3
|
|
"""Agent for auditing zero trust controls on SaaS applications via Microsoft Graph API."""
|
|
|
|
import requests
|
|
import json
|
|
import argparse
|
|
from datetime import datetime, timezone
|
|
|
|
GRAPH_API = "https://graph.microsoft.com/v1.0"
|
|
|
|
|
|
def get_token(tenant_id, client_id, client_secret):
|
|
"""Acquire OAuth2 token for Microsoft Graph API."""
|
|
url = f"https://login.microsoftonline.com/{tenant_id}/oauth2/v2.0/token"
|
|
data = {"grant_type": "client_credentials", "client_id": client_id,
|
|
"client_secret": client_secret, "scope": "https://graph.microsoft.com/.default"}
|
|
resp = requests.post(url, data=data, timeout=30)
|
|
resp.raise_for_status()
|
|
token = resp.json()["access_token"]
|
|
print("[*] Authenticated to Microsoft Graph API")
|
|
return {"Authorization": f"Bearer {token}"}
|
|
|
|
|
|
def list_conditional_access_policies(headers):
|
|
"""List Entra ID conditional access policies for SaaS apps."""
|
|
url = f"{GRAPH_API}/identity/conditionalAccess/policies"
|
|
resp = requests.get(url, headers=headers, timeout=30)
|
|
resp.raise_for_status()
|
|
policies = resp.json().get("value", [])
|
|
findings = []
|
|
print(f"\n[*] Conditional Access Policies: {len(policies)}")
|
|
for p in policies:
|
|
state = p.get("state", "disabled")
|
|
conditions = p.get("conditions", {})
|
|
grant_controls = p.get("grantControls", {})
|
|
mfa_required = "mfa" in str(grant_controls.get("builtInControls", [])).lower()
|
|
print(f" [{'+' if state == 'enabled' else '-'}] {p['displayName']} "
|
|
f"(state={state}, MFA={'Yes' if mfa_required else 'No'})")
|
|
if state == "enabled" and not mfa_required:
|
|
findings.append({"policy": p["displayName"], "issue": "No MFA required",
|
|
"severity": "HIGH"})
|
|
return policies, findings
|
|
|
|
|
|
def list_enterprise_apps(headers):
|
|
"""List enterprise applications (service principals) for shadow IT discovery."""
|
|
url = f"{GRAPH_API}/servicePrincipals"
|
|
params = {"$top": 100, "$select": "displayName,appId,accountEnabled,signInAudience"}
|
|
resp = requests.get(url, headers=headers, params=params, timeout=30)
|
|
resp.raise_for_status()
|
|
apps = resp.json().get("value", [])
|
|
print(f"\n[*] Enterprise Applications: {len(apps)}")
|
|
third_party = [a for a in apps if a.get("signInAudience") != "AzureADMyOrg"]
|
|
print(f" Third-party apps: {len(third_party)}")
|
|
for a in third_party[:10]:
|
|
print(f" {a['displayName']} (enabled={a.get('accountEnabled', '?')})")
|
|
return apps
|
|
|
|
|
|
def check_oauth_app_consents(headers):
|
|
"""Audit OAuth2 permission grants for overprivileged consents."""
|
|
url = f"{GRAPH_API}/oauth2PermissionGrants"
|
|
resp = requests.get(url, headers=headers, timeout=30)
|
|
resp.raise_for_status()
|
|
grants = resp.json().get("value", [])
|
|
findings = []
|
|
for g in grants:
|
|
scope = g.get("scope", "")
|
|
if any(perm in scope for perm in ["Mail.ReadWrite", "Files.ReadWrite.All",
|
|
"Directory.ReadWrite.All", "User.ReadWrite.All"]):
|
|
findings.append({
|
|
"clientId": g.get("clientId"), "scope": scope,
|
|
"consentType": g.get("consentType"), "severity": "HIGH",
|
|
})
|
|
print(f"\n[*] OAuth grants: {len(grants)} total, {len(findings)} overprivileged")
|
|
for f in findings[:5]:
|
|
print(f" [!] {f['clientId']}: {f['scope'][:80]}")
|
|
return findings
|
|
|
|
|
|
def check_sign_in_risk(headers, days=7):
|
|
"""Check risky sign-ins to SaaS applications."""
|
|
url = f"{GRAPH_API}/identityProtection/riskyUsers"
|
|
params = {"$filter": "riskState eq 'atRisk'", "$top": 50}
|
|
resp = requests.get(url, headers=headers, params=params, timeout=30)
|
|
if resp.status_code == 200:
|
|
users = resp.json().get("value", [])
|
|
print(f"\n[*] Risky users: {len(users)}")
|
|
for u in users[:10]:
|
|
print(f" [!] {u.get('userDisplayName', 'N/A')} - risk: {u.get('riskLevel')}")
|
|
return users
|
|
return []
|
|
|
|
|
|
def generate_report(policies, ca_findings, oauth_findings, risky_users, output_path):
|
|
"""Generate SaaS zero trust audit report."""
|
|
report = {
|
|
"audit_date": datetime.now(timezone.utc).isoformat(),
|
|
"summary": {
|
|
"conditional_access_policies": len(policies),
|
|
"ca_findings": len(ca_findings),
|
|
"overprivileged_oauth": len(oauth_findings),
|
|
"risky_users": len(risky_users),
|
|
},
|
|
"ca_findings": ca_findings,
|
|
"oauth_findings": oauth_findings[:20],
|
|
"risky_users": risky_users[:20],
|
|
}
|
|
with open(output_path, "w") as f:
|
|
json.dump(report, f, indent=2, default=str)
|
|
print(f"\n[*] Report saved to {output_path}")
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description="SaaS Zero Trust Audit Agent")
|
|
parser.add_argument("action", choices=["ca-policies", "apps", "oauth", "risk", "full-audit"])
|
|
parser.add_argument("--tenant-id", required=True)
|
|
parser.add_argument("--client-id", required=True)
|
|
parser.add_argument("--client-secret", required=True)
|
|
parser.add_argument("-o", "--output", default="saas_zt_audit.json")
|
|
args = parser.parse_args()
|
|
|
|
headers = get_token(args.tenant_id, args.client_id, args.client_secret)
|
|
if args.action == "ca-policies":
|
|
list_conditional_access_policies(headers)
|
|
elif args.action == "apps":
|
|
list_enterprise_apps(headers)
|
|
elif args.action == "oauth":
|
|
check_oauth_app_consents(headers)
|
|
elif args.action == "risk":
|
|
check_sign_in_risk(headers)
|
|
elif args.action == "full-audit":
|
|
policies, ca_f = list_conditional_access_policies(headers)
|
|
list_enterprise_apps(headers)
|
|
oauth_f = check_oauth_app_consents(headers)
|
|
risky = check_sign_in_risk(headers)
|
|
generate_report(policies, ca_f, oauth_f, risky, args.output)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|