Skip to content

aniket2348823/vigilagent / .agents/skills/analyzing-office365-audit-logs-for-compromise

analyzing-office365-audit-logs-for-compromise

Verified

Parse Office 365 Unified Audit Logs via Microsoft Graph API to detect email forwarding rule creation, inbox delegation, suspicious OAuth app grants, and other indicators of account compromise.

FileKindSize
LICENSEother11.0 KB
SKILL.mdskill md2.2 KB
references/api-reference.mdmarkdown2.8 KB
scripts/agent.pyscript9.8 KB

scripts/agent.py

View on GitHub
#!/usr/bin/env python3
"""Agent for analyzing Office 365 audit logs for compromise indicators via Microsoft Graph."""

import json
import argparse
from datetime import datetime, timedelta

try:
    import msal
except ImportError:
    msal = None

try:
    import requests
except ImportError:
    requests = None

GRAPH_BASE = "https://graph.microsoft.com/v1.0"

SUSPICIOUS_OPERATIONS = [
    "New-InboxRule", "Set-InboxRule", "Set-Mailbox",
    "Add-MailboxPermission", "Set-MailboxJunkEmailConfiguration",
    "Set-OwaMailboxPolicy", "New-TransportRule",
    "Add-RecipientPermission", "Set-TransportRule",
    "UpdateInboxRules", "Set-MailboxAutoReplyConfiguration",
]


def get_access_token(tenant_id, client_id, client_secret):
    """Authenticate via MSAL client credentials flow."""
    if not msal:
        return None
    authority = f"https://login.microsoftonline.com/{tenant_id}"
    app = msal.ConfidentialClientApplication(
        client_id, authority=authority, client_credential=client_secret
    )
    result = app.acquire_token_for_client(scopes=["https://graph.microsoft.com/.default"])
    if "access_token" in result:
        return result["access_token"]
    raise RuntimeError(f"Auth failed: {result.get('error_description', result.get('error'))}")


def graph_get(token, endpoint, params=None):
    """Paginated GET against Microsoft Graph API."""
    headers = {"Authorization": f"Bearer {token}", "Content-Type": "application/json"}
    url = f"{GRAPH_BASE}{endpoint}"
    all_items = []
    while url:
        resp = requests.get(url, headers=headers, params=params, timeout=30)
        resp.raise_for_status()
        data = resp.json()
        all_items.extend(data.get("value", []))
        url = data.get("@odata.nextLink")
        params = None
    return all_items


def query_audit_logs(token, days=7):
    """Query Unified Audit Log for suspicious mail operations."""
    since = (datetime.utcnow() - timedelta(days=days)).strftime("%Y-%m-%dT%H:%M:%SZ")
    events = []
    for operation in SUSPICIOUS_OPERATIONS:
        filter_str = (
            f"activityDisplayName eq '{operation}' "
            f"and activityDateTime ge {since}"
        )
        try:
            logs = graph_get(token, "/auditLogs/directoryAudits",
                             params={"$filter": filter_str, "$top": "100"})
            for log in logs:
                initiated = log.get("initiatedBy", {}).get("user", {})
                events.append({
                    "operation": operation,
                    "timestamp": log.get("activityDateTime"),
                    "result": log.get("result"),
                    "user": initiated.get("userPrincipalName"),
                    "ip_address": initiated.get("ipAddress"),
                    "target": [t.get("displayName") for t in log.get("targetResources", [])],
                    "details": log.get("additionalDetails"),
                })
        except Exception:
            pass
    events.sort(key=lambda x: x.get("timestamp", ""), reverse=True)
    return events


def check_inbox_rules(token, user_id):
    """List inbox rules for a specific mailbox and flag forwarding rules."""
    rules = graph_get(token, f"/users/{user_id}/mailFolders/inbox/messageRules")
    findings = []
    for rule in rules:
        is_forwarding = False
        forward_to = []
        actions = rule.get("actions", {})
        if actions.get("forwardTo"):
            is_forwarding = True
            forward_to = [r.get("emailAddress", {}).get("address", "")
                          for r in actions["forwardTo"]]
        if actions.get("forwardAsAttachmentTo"):
            is_forwarding = True
            forward_to += [r.get("emailAddress", {}).get("address", "")
                           for r in actions["forwardAsAttachmentTo"]]
        if actions.get("redirectTo"):
            is_forwarding = True
            forward_to += [r.get("emailAddress", {}).get("address", "")
                           for r in actions["redirectTo"]]
        delete_after = actions.get("delete", False)
        mark_read = actions.get("markAsRead", False)

        risk = 0
        if is_forwarding:
            risk += 40
        if delete_after:
            risk += 25
        if mark_read and is_forwarding:
            risk += 15
        external = [f for f in forward_to if f and not f.endswith(user_id.split("@")[-1])]
        if external:
            risk += 20

        findings.append({
            "rule_id": rule.get("id"),
            "display_name": rule.get("displayName"),
            "enabled": rule.get("isEnabled"),
            "is_forwarding": is_forwarding,
            "forward_to": forward_to,
            "external_forwards": external,
            "delete_after_forward": delete_after,
            "mark_as_read": mark_read,
            "conditions": rule.get("conditions"),
            "risk_score": min(risk, 100),
            "risk_level": "CRITICAL" if risk >= 70 else "HIGH" if risk >= 50 else "MEDIUM" if risk >= 25 else "LOW",
        })
    return findings


def check_mailbox_forwarding(token, user_id):
    """Check mailbox-level SMTP forwarding configuration."""
    try:
        headers = {"Authorization": f"Bearer {token}"}
        resp = requests.get(
            f"{GRAPH_BASE}/users/{user_id}/mailboxSettings",
            headers=headers, timeout=30
        )
        resp.raise_for_status()
        settings = resp.json()
        auto_reply = settings.get("automaticRepliesSetting", {})
        return {
            "user": user_id,
            "auto_reply_enabled": auto_reply.get("status") != "disabled",
            "auto_reply_external_audience": auto_reply.get("externalAudience"),
            "language": settings.get("language", {}).get("locale"),
            "timezone": settings.get("timeZone"),
        }
    except Exception as e:
        return {"user": user_id, "error": str(e)}


def check_oauth_grants(token):
    """Check for suspicious OAuth application consent grants."""
    grants = graph_get(token, "/oauth2PermissionGrants")
    high_risk_scopes = {"Mail.Read", "Mail.ReadWrite", "Mail.Send",
                        "Files.ReadWrite.All", "MailboxSettings.ReadWrite"}
    suspicious = []
    for g in grants:
        scopes = g.get("scope", "").split()
        risky = [s for s in scopes if s in high_risk_scopes]
        if risky:
            suspicious.append({
                "client_id": g.get("clientId"),
                "consent_type": g.get("consentType"),
                "principal_id": g.get("principalId"),
                "high_risk_scopes": risky,
                "all_scopes": scopes,
                "risk_score": min(len(risky) * 20, 100),
            })
    suspicious.sort(key=lambda x: x["risk_score"], reverse=True)
    return suspicious


def full_audit(token, users=None, days=7):
    """Run comprehensive O365 compromise audit."""
    audit_events = query_audit_logs(token, days)
    oauth_findings = check_oauth_grants(token)

    inbox_findings = []
    forwarding_findings = []
    if users:
        for user in users:
            rules = check_inbox_rules(token, user)
            inbox_findings.extend([{**r, "mailbox": user} for r in rules])
            fwd = check_mailbox_forwarding(token, user)
            forwarding_findings.append(fwd)

    suspicious_rules = [r for r in inbox_findings if r.get("is_forwarding")]
    external_forwards = [r for r in inbox_findings if r.get("external_forwards")]

    return {
        "audit_type": "Office 365 Compromise Analysis",
        "timestamp": datetime.utcnow().isoformat(),
        "lookback_days": days,
        "summary": {
            "suspicious_audit_events": len(audit_events),
            "oauth_high_risk_grants": len(oauth_findings),
            "forwarding_rules_found": len(suspicious_rules),
            "external_forwarding_rules": len(external_forwards),
            "mailboxes_scanned": len(users) if users else 0,
        },
        "audit_events": audit_events[:25],
        "oauth_findings": oauth_findings[:15],
        "forwarding_rules": suspicious_rules[:20],
        "mailbox_settings": forwarding_findings[:20],
    }


def main():
    parser = argparse.ArgumentParser(description="Office 365 Audit Log Compromise Analysis Agent")
    parser.add_argument("--tenant-id", required=True, help="Azure AD tenant ID")
    parser.add_argument("--client-id", required=True, help="App registration client ID")
    parser.add_argument("--client-secret", required=True, help="App client secret")
    sub = parser.add_subparsers(dest="command")
    p_audit = sub.add_parser("audit-logs", help="Query suspicious audit events")
    p_audit.add_argument("--days", type=int, default=7)
    p_rules = sub.add_parser("inbox-rules", help="Check inbox rules for forwarding")
    p_rules.add_argument("--user", required=True, help="User principal name")
    p_fwd = sub.add_parser("forwarding", help="Check mailbox forwarding settings")
    p_fwd.add_argument("--user", required=True)
    sub.add_parser("oauth", help="Check OAuth consent grants")
    p_full = sub.add_parser("full", help="Full compromise audit")
    p_full.add_argument("--users", nargs="+", help="User principal names to scan")
    p_full.add_argument("--days", type=int, default=7)
    args = parser.parse_args()

    token = get_access_token(args.tenant_id, args.client_id, args.client_secret)

    if args.command == "audit-logs":
        result = query_audit_logs(token, args.days)
    elif args.command == "inbox-rules":
        result = check_inbox_rules(token, args.user)
    elif args.command == "forwarding":
        result = check_mailbox_forwarding(token, args.user)
    elif args.command == "oauth":
        result = check_oauth_grants(token)
    elif args.command == "full" or args.command is None:
        result = full_audit(token, getattr(args, "users", None), getattr(args, "days", 7))
    else:
        parser.print_help()
        return
    print(json.dumps(result, indent=2, default=str))


if __name__ == "__main__":
    main()
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall