Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
4,916 detections
Filters
Last updated
All Time
Detection languages
2,236
1,064
466
410
326
Contributors
725
665
535
485
271
Categories
1,085
870
805
765
576
Platforms
1,843
1,022
856
814
656
Products / Services
988
596
477
456
383
MITRE Techniques
18,115
15,635
12,577
8,388
6,215
4,916
CVEs
37
17
17
15
12
IDS Classtypes
196
73
47
29
16
IDS Protocols
301
27
24
22
6
This rule detects unauthorized exploitation attempts against Gitea environments vulnerable to CVE-2026-20896. When Gitea is configured with REVERSE_PROXY_TRUSTED_PROXIES set to permit all sources, an attacker can bypass authentication by injecting the 'X-WEBAUTH-USER: admin' header in HTTP requests. This allows the attacker to gain administrative privileges and perform unauthorized actions, such as creating new admin accounts or accessing restricted API endpoints.
This rule monitors for security alerts related to Azure AI services, specifically detecting credential theft attempts and LLM jailbreak attempts that have been blocked or detected by Azure's built-in content filtering mechanisms.
Detects Parameter-to-Prompt (P2P) injection attempts targeting Microsoft 365 Copilot by identifying suspicious patterns in URL query parameters. The rule monitors AuditLogs for queries containing HTML tags, encoded payloads, or known malicious markers typical of prompt injection exploits against search functionalities (CVE-2026-42824).
This rule detects potential lateral movement or account compromise by identifying instances where a single user account successfully logs into four or more unique computer systems via Remote Desktop (Logon Type 10) within a 30-minute window.
Detects a suspicious sequence of events involving the creation of an .amazonq/mcp.json file shortly after a git clone operation, followed by an AI coding assistant (Amazon Q, VS Code, Cursor, Windsurf) spawning runtime interpreters (node, npx, python). This behavior aligns with potential exploitation attempts against AI-assisted development environments, such as CVE-2026-12957, where malicious configuration files are used to trigger unauthorized code execution.
Detects the exploitation of CVE-2026-20253, where the Flowise application fails to correctly sanitize or validate environment variables. By using a case-insensitive bypass (e.g., lowercase 'node_options'), an attacker can cause a spawned Node.js child process to honor the variable and execute unauthorized JavaScript files via the '--require' command-line argument.
Detects the creation of .amazonq/mcp.json or mcp.json files within repository directories following a git clone operation. This behavior is associated with the exploitation of vulnerabilities (such as CVE-2026-12957) where attackers use malicious job-interview repositories to trigger automatic execution of arbitrary MCP (Model Context Protocol) servers in the Amazon Q Developer extension in VS Code, potentially leading to unauthorized credential or sensitive data access.
Detects the spawning of potential Model Context Protocol (MCP) server processes immediately following or by package installation commands (npm, pip). This behavior is characteristic of malicious packages or typosquatted dependencies attempting to register or execute a malicious MCP server configuration upon installation in a developer environment.
Detects incoming phishing emails sent from personal Gmail accounts that attempt to impersonate law firms by using legal-themed subjects (e.g., lawsuits, subpoenas, cease and desist). The rule specifically identifies the use of URL shorteners commonly used to obfuscate credential-harvesting destinations. The detection relies on email metadata and subject line patterns to identify potential spearphishing attempts.
Detects successful exploitation of the CVE-2026-48558 OIDC authentication bypass vulnerability in SimpleHelp. The rule monitors for a forged OIDC response processed by the OIDCAuthenticator followed immediately by the registration of a new technician account marked with the [New Anon] identifier, indicating identity verification failure and unauthorized privileged account creation.
Detects instances where AI coding agents, specifically Claude or Node-based environments, initiate secondary processes like git clone, package installation (pip, npm), or script execution. This behavior often suggests the agent is autonomously pulling and executing code from external repositories without explicit human approval, posing a significant risk of supply chain compromise.
Detects instances where a user successfully authenticates from a mobile device (Android or iPhone) that exhibits a geographic mismatch (geo_country vs registered_country), and subsequently performs a sensitive credential or MFA-related action (enrollment or reset) from a different device within one hour. This pattern is indicative of a potential session hijacking or credential compromise where an attacker attempts to establish persistence by enrolling their own MFA device.
Detects the use of legitimate SSH clients (ssh.exe, plink.exe) to initiate outbound network connections to internal IP ranges on port 22. This activity is consistent with Obsidian Sandstorm lateral movement tactics, where compromised credentials are used to establish SSH tunnels for internal network pivoting.
Detects potential exploitation of the Zerologon vulnerability (CVE-2020-1472) by correlating computer account password changes (Event ID 4742) with preceding anonymous or null SID logons (Event ID 4624) where the host identifier aligns. The rule identifies anomalous account modifications that occur within a short time window following an anonymous authentication request on the same host.
This rule detects potential exploitation attempts targeting the BeyondTrust Privileged Remote Access API by identifying unauthenticated POST requests directed at specific sensitive API endpoints (e.g., /api/, /remote-support/). It flags high-frequency or multi-path request patterns indicative of automated vulnerability scanning or exploitation attempts related to CVE-2026-1731, where the absence of authorization headers is a key indicator of unauthorized access.
This rule detects interactive logons (LogonType 10) where a single account from the same domain logs into multiple distinct devices. This could indicate an attacker using compromised credentials to move laterally within the network or access multiple systems.
This rule detects potential bulk phishing email campaigns by identifying a high volume of emails originating from external senders (not @company.com) with suspicious subject lines containing keywords like 'action required', 'verify account', or 'confirm identity'. The rule triggers if at least 50 such emails are sent to 20 or more unique recipients within a one-hour window.
This rule detects when a single user principal name (UPN) attempts five or more sign-ins from the same IP address within a one-hour window, and each sign-in is flagged with a 'RiskDetail' other than 'none' but still results in a 'Success' status. This could indicate an attacker attempting to bypass security controls or an account compromise where the attacker is repeatedly trying to access the account from a consistent location, possibly with different credentials or methods, until successful.
Detects incoming web requests to GCP Cloud Run services that exhibit patterns associated with Server-Side Request Forgery (SSRF) exploitation. The rule monitors for URLs containing GitHub blob paths combined with path traversal sequences ('//') that point to sensitive internal system paths such as /proc/, /etc/, or /var/task/.
This rule detects potentially malicious HTTP requests targeted at Google Cloud Run services that attempt to access restricted or internal application paths, such as '/app/' or '/var/task/'. These patterns often indicate attempts to probe for or exploit vulnerable web application files, internal scripts, or configuration files (e.g., 'graph_api.py', 'graph.zip') commonly found in serverless environments.
Detects potential exploitation attempts against management interfaces of SOHO network devices (Cisco, Ubiquiti, Draytek) by monitoring for suspicious web request activity, specifically targeting paths like /admin, /cgi-bin, or /setup.cgi. The rule identifies high-volume or systematic scan patterns from external sources, which may indicate automated exploitation attempts (e.g., related to CVE-2026-35616).
Page 30 of 235