Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
4,230 detections
Filters
Last updated
All Time
Detection languages
2,175
1,018
354
260
212
Contributors
1,013
818
688
574
212
Categories
20,456
11,839
5,877
5,156
4,865
4,230
Platforms
2,933
889
585
248
198
Products / Services
2,239
1,453
841
416
344
MITRE Techniques
1,164
1,087
870
798
439
CVEs
22
20
20
18
17
IDS Classtypes
927
408
222
146
104
IDS Protocols
1,914
105
70
47
6
Detects network activity associated with the 'ClickFix' social engineering campaign, which commonly targets users with fake error messages to trick them into executing malicious commands. The rule identifies communication with known malicious infrastructure (domains and IP addresses) found in CommonSecurityLog events, while implementing filters to exclude common security scanners and deduplicating per-host alerts.
Detects network activity associated with the 'ClickFix' social engineering campaign, which commonly targets users with fake error messages to trick them into executing malicious commands. The rule identifies communication with known malicious infrastructure (domains and IP addresses) found in CommonSecurityLog events, while implementing filters to exclude common security scanners and deduplicating per-host alerts.
This rule detects potentially malicious Command and Control (C2) traffic utilizing DNS over HTTPS (DoH) to interact with public resolvers like Cloudflare, Google, and Quad9. It identifies both high-frequency, repeated direct-to-resolver TLS sessions and specific HTTP requests to these resolvers that resolve domains associated with known C2 infrastructure, such as the ClickFix campaign.
Detects NeedyMantis' initial C2 beacon: an HTTP GET to /library/zip/ on port 443, anchored to the URI root, carrying a client-originated 'Set-Cookie:' header (clients normally send 'Cookie:', never 'Set-Cookie:' -- this direction reversal is the core anomaly and is near-impossible for legitimate traffic) alongside an explicit 'Upgrade: websocket' header. Matching the literal header text (not just header-name presence) and anchoring the URI removes generic WebSocket apps and any unrelated path containing '/library/zip/' as a substring.
Hunts telemetry for published NeedyMantis indicators: three known SHA-256 hashes (WinSparkle.dll first-stage loader and its encrypted archive, plus the older libcurl archive), the C2 domain corp.tripswithengine[.]com, and the C2 URL path /library/zip/ on that domain. Domain/URL matching parses the actual host out of RemoteUrl and requires an EXACT match (not substring 'has/contains'), so a different domain that merely contains 'corp.tripswithengine.com' as a suffix of a longer label (e.g. 'notcorp.tripswithengine.com') or as a prefix of an unrelated domain (e.g. 'corp.tripswithengine.com.evil.net') cannot match. No malicious IP address has been published for this campaign, so IP-based matching is intentionally not included.
This rule detects a multi-stage phishing redirection chain where a user visits a known survey platform (t.maze.co) followed immediately by a request to a Cloudflare Workers landing page (.workers.dev), a common pattern used by adversaries to obfuscate malicious delivery endpoints.
Detects high-frequency web requests from a single source IP containing URI patterns associated with known vulnerability exploits (e.g., directory traversal, remote code execution, or path injection). This rule identifies potential mass exploitation attempts against public-facing web applications by monitoring firewall logs for a threshold of suspicious URI activity.
Detects web requests targeting common LLM/AI inference API endpoints containing strings indicative of prompt injection attempts, such as attempts to bypass filters, exfiltrate system instructions, or force model behavior changes (e.g., DAN mode).
Detects large outbound HTTP POST requests to known public AI-inference APIs and paste services. This pattern is potentially indicative of data exfiltration or sensitive information exposure where attackers leverage third-party services to bypass traditional outbound egress filtering.
Detects outbound HTTP requests traversing a proxy where the User-Agent string contains structurally consistent but algorithmically randomized components. This pattern is characteristic of AI-generated C2 client code attempting to create plausible yet non-standard browser signatures for each beacon. Detection relies on identifying User-Agent strings that follow a standard browser template (e.g., Chrome/Safari) but include anomalous hex/alphanumeric tokens in the version fields.
Detects web requests targeting sensitive Jenkins/CI-CD administrative and build-triggering endpoints, such as /script, /manage, and /job/*/build. These paths are high-value targets for reconnaissance and automated exploitation efforts. This rule identifies potentially malicious activity by monitoring for access to these specific paths, which should be correlated with frequency and source data to differentiate automated scanning from authorized administrative or system activities.
Detects unauthenticated POST requests to the TorchServe Management API '/models' endpoint containing a 'url=' parameter. This pattern is indicative of an attempt to perform remote model registration, which could be leveraged for Remote Code Execution (RCE) on the server hosting the TorchServe service.
This rule detects unauthorized and potentially unauthenticated POST requests directed at the completion API endpoints (/api/v1/generate and /run/textgen) of the Oobabooga text-generation-webui, which is commonly used to host Large Language Models (LLMs). Such access, if not secured, could allow an external actor to interact with the model or potentially exploit the underlying infrastructure.
This rule detects POST requests to the '/v2/repository/index' endpoint of a Triton Inference Server. This endpoint is part of the Model Repository control API and is typically used to manage or inspect models. Unauthorized access to this API can be indicative of reconnaissance or exploitation attempts against the inference server.
This rule detects unauthorized and unauthenticated attempts to access the Ray Dashboard API '/api/jobs' endpoint. An absence of an Authorization header in these requests indicates potential reconnaissance or exploitation attempts against an exposed Ray cluster management interface.
Detects incoming HTTP GET requests to common Jupyter Notebook endpoints ('/tree' or '/api/kernels') that do not include a token in the URI. This behavior is indicative of an adversary attempting to interact with or explore an insecurely configured Jupyter Notebook server that lacks authentication.
Detects high-frequency requests from automated scripting tools like 'python-requests', 'aiohttp', or 'Scrapy' targeting sensitive directories or files often associated with administrative panels or application configurations (e.g., admin, .env, wp-admin). The rule triggers when a threshold of 10 requests is reached within a 60-second window, indicating potential automated vulnerability or information gathering activity.
Detects multiple inbound HTTP requests targeting known vulnerable application paths (e.g., cgi-bin, .env files, actuator gateways, path traversal, WordPress plugins, PHP eval files, and various server administration consoles) from a single source within a short timeframe. This behavior is indicative of an automated scanner or exploit chain attempting to identify and compromise public-facing applications.
Detects unauthenticated POST requests directed at the '/v1/completions' endpoint of a vLLM server. This behavior is indicative of potential unauthorized access or exploitation attempts against an exposed LLM inference API that lacks authentication controls.
Detects unauthenticated HTTP POST requests targeting the '/v1/graphql' endpoint of a Weaviate vector database. This pattern is indicative of reconnaissance or exploitation attempts against exposed Weaviate API instances.
Detects network traffic associated with Adversary-in-the-Middle (AiTM) phishing kits that impersonate Microsoft and Google login pages. The rules identify the presence of specific HTML elements (like Cloudflare Turnstile anti-bot gates) and authentication session cookie headers served from non-authorized/non-official domains, indicating a reverse-proxy phishing attack.
Page 19 of 202