Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
381 detections
Filters
Last updated
All Time
Detection languages
380
1
Contributors
160
79
64
38
20
Categories
220
190
146
124
101
Platforms
325
56
56
48
11
Products / Services
362
188
33
20
19
MITRE Techniques
140
73
56
53
51
CVEs
12
9
9
5
5
IDS Classtypes
1,937
503
461
381
236
IDS Protocols
189
154
25
4
4
Detects potential unauthenticated Remote Code Execution (RCE) exploitation attempts targeting the IKEv2/ISAKMP VPN service on WatchGuard Firebox devices (CVE-2025-14733). The rule monitors UDP traffic on standard IKE ports 500 and 4500, looking for specific payload patterns and elevated data sizes consistent with exploitation attempts, coupled with a threshold to reduce noise.
Detects unauthenticated HTTP POST requests targeting the FortiGate administrative API (/api/v2/), which may indicate exploitation attempts related to CVE-2025-25249. The rule specifically looks for requests lacking authentication headers (Cookie or Authorization).
Detects an attempt to exploit CVE-2026-24858 in FortiCloud SSO by sending a crafted SAML authentication request that lacks a signature. This vulnerability allows an attacker to bypass authentication by providing an unsigned SAML response.
Detects incoming HTTP POST requests to the GitLab '/import' endpoint that attempt to access the internal cloud metadata service (169.254.169.254), indicating an exploitation attempt for vulnerabilities such as CVE-2021-22175 or CVE-2021-39935.
Detects unauthorized or suspicious POST requests to the Cisco Firepower Management Center (FMC) web interface that lack standard authentication headers (Cookie, Authorization, or X-Api-Token). This pattern is consistent with exploit attempts targeting CVE-2026-20079.
Detects a multi-stage authentication bypass attempt against LiteLLM MCP endpoints. The rule tracks an initial malformed Authorization Bearer token request to the /mcp/ endpoint, followed by a server response returning a 200 OK status code accompanied by an 'mcp-session-id' header, confirming the exploitation of CVE-2026-59822.
Detects possible unauthenticated exploitation attempts against CVE-2025-53521 (F5 BIG-IP APM RCE). Tightened from a generic 'any POST without a cookie on 443/8443' match (extremely broad, alerts on ordinary cookie-less API/health-check traffic) to POST requests specifically targeting F5 APM access-policy processing endpoints (/my.policy, /vdesk/) that lack a session cookie, since the CVE requires an access policy configured on the virtual server and legitimate APM sessions carry a session cookie. Added detection_filter (3 hits/60s per source) to suppress single incidental matches and prioritize repeated attempts. Note: public reporting for this CVE does not include the exact exploit URI/payload, so this is a behavioral/namespace heuristic, not a payload-specific signature — validate against your own APM traffic baseline and update once exploit specifics are published.
Detects possible unauthenticated exploitation attempts against CVE-2025-53521 (F5 BIG-IP APM RCE). Tightened from a generic 'any POST without a cookie on 443/8443' match (extremely broad, alerts on ordinary cookie-less API/health-check traffic) to POST requests specifically targeting F5 APM access-policy processing endpoints (/my.policy, /vdesk/) that lack a session cookie, since the CVE requires an access policy configured on the virtual server and legitimate APM sessions carry a session cookie. Added detection_filter (3 hits/60s per source) to suppress single incidental matches and prioritize repeated attempts. Note: public reporting for this CVE does not include the exact exploit URI/payload, so this is a behavioral/namespace heuristic, not a payload-specific signature — validate against your own APM traffic baseline and update once exploit specifics are published.
Detects incoming HTTP POST requests to the GitLab '/import' endpoint that attempt to access the internal cloud metadata service (169.254.169.254), indicating an exploitation attempt for vulnerabilities such as CVE-2021-22175 or CVE-2021-39935.
Detects incoming HTTP POST requests to the GitLab '/import' endpoint that attempt to access the internal cloud metadata service (169.254.169.254), indicating an exploitation attempt for vulnerabilities such as CVE-2021-22175 or CVE-2021-39935.
Detects anomalous network connection attempts using DCOM (port 135) or WinRM (ports 5985/5986) that originate from external networks or exceed established internal baseline frequencies. This pattern is often indicative of lateral movement or remote administration abuse via WMI or WinRM.
Detects anomalous network connection attempts using DCOM (port 135) or WinRM (ports 5985/5986) that originate from external networks or exceed established internal baseline frequencies. This pattern is often indicative of lateral movement or remote administration abuse via WMI or WinRM.
Detects an attempt to exploit CVE-2026-24858 in FortiCloud SSO by sending a crafted SAML authentication request that lacks a signature. This vulnerability allows an attacker to bypass authentication by providing an unsigned SAML response.
This rule monitors incoming HTTP requests for the presence of the Base64-encoded string 'rO0AB', which is the standard header for serialized Java objects. The presence of this header in an HTTP request body is a strong indicator of a Java deserialization exploitation attempt.
This rule detects common patterns associated with Expression Language (EL) injection attacks in HTTP traffic. These patterns typically attempt to invoke Java reflection, runtime execution, or process creation through vulnerable applications utilizing templating engines or expression evaluators (e.g., OGNL, Spring Expression Language). Attackers use these to achieve remote code execution (RCE).
Detects a multi-stage authentication bypass attempt against LiteLLM MCP endpoints. The rule tracks an initial malformed Authorization Bearer token request to the /mcp/ endpoint, followed by a server response returning a 200 OK status code accompanied by an 'mcp-session-id' header, confirming the exploitation of CVE-2026-59822.
Detects a multi-stage authentication bypass attempt against LiteLLM MCP endpoints. The rule tracks an initial malformed Authorization Bearer token request to the /mcp/ endpoint, followed by a server response returning a 200 OK status code accompanied by an 'mcp-session-id' header, confirming the exploitation of CVE-2026-59822.
Detects inbound network connections from known IP addresses associated with the StyleSmuggler malware distribution, which is used to inject malicious scripts into web applications.
Detects attempts to exploit the Log4j vulnerability (CVE-2021-44228) by identifying JNDI lookups (ldap, rmi, http, ldaps, dns) in inbound HTTP traffic. These lookups are a hallmark of the Log4Shell attack vector, allowing for remote code execution.
Detects possible unauthenticated exploitation attempts against CVE-2025-53521 (F5 BIG-IP APM RCE). Tightened from a generic 'any POST without a cookie on 443/8443' match (extremely broad, alerts on ordinary cookie-less API/health-check traffic) to POST requests specifically targeting F5 APM access-policy processing endpoints (/my.policy, /vdesk/) that lack a session cookie, since the CVE requires an access policy configured on the virtual server and legitimate APM sessions carry a session cookie. Added detection_filter (3 hits/60s per source) to suppress single incidental matches and prioritize repeated attempts. Note: public reporting for this CVE does not include the exact exploit URI/payload, so this is a behavioral/namespace heuristic, not a payload-specific signature — validate against your own APM traffic baseline and update once exploit specifics are published.
Detects inbound network traffic originating from IP addresses known to be associated with an exploit campaign targeting the Super Forms plugin (CVE-2026-14894). The rule acts as an indicator of compromise (IoC) match for potential exploitation attempts directed at internal network infrastructure.
Page 7 of 19