Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,733 detections
Filters
Last updated
All Time
Detection languages
909
387
164
127
68
Contributors
479
417
201
177
61
Categories
20,440
11,814
5,871
5,145
4,864
1,733
Platforms
822
436
329
274
206
Products / Services
498
378
118
96
79
MITRE Techniques
433
327
225
196
196
CVEs
12
12
10
10
6
IDS Classtypes
1
1
IDS Protocols
2
Detects the sensitive project-wide CI/CD variable API_TOKEN being exposed to job scopes other than the gitlab-triage automation job that is its only intended consumer, expanding the blast radius of the credential.
Detects high-velocity publication of Ruby packages by an account that uses naming conventions related to 'OAI' or a known-malicious contact email. This behavior is indicative of a supply chain attack involving the automated deployment of numerous malicious packages to the RubyGems ecosystem.
Detects high-velocity publication of Ruby packages by an account that uses naming conventions related to 'OAI' or a known-malicious contact email. This behavior is indicative of a supply chain attack involving the automated deployment of numerous malicious packages to the RubyGems ecosystem.
Detects N-able N-central Take Control remote session start/end events in the Windows Application log where the viewer identity is the default N-central support account 'MSP Support' / mspsupport@n-able.com, or the source/viewer IP matches known attacker infrastructure associated with active exploitation of CVE-2026-18577.
Detects successful authentication attempts to a LiteLLM proxy instance using the hardcoded default master key 'sk-1234'. This activity indicates potential exploitation of misconfigured instances where default credentials remain unchanged.
Detects successful authentication attempts to a LiteLLM proxy instance using the hardcoded default master key 'sk-1234'. This activity indicates potential exploitation of misconfigured instances where default credentials remain unchanged.
Detects successful authentication attempts to a LiteLLM proxy instance using the hardcoded default master key 'sk-1234'. This activity indicates potential exploitation of misconfigured instances where default credentials remain unchanged.
Detects sessions using shared Grafana service account tokens that perform both read-only and action-capable tool operations without confirmed prior authentication. This behavior is indicative of an unauthenticated or spoofed session leveraging a service account's excessive permissions, bypassing standard authentication flows.
Detects incoming Grafana API requests directed at vulnerable mcp-grafana server instances identified by an outdated version (pre-1.1.0). This rule correlates host-level version discovery with application-layer logs containing suspicious 'grafana_api_request' method calls, which may indicate an attempt to exploit Server-Side Request Forgery (SSRF) vulnerabilities.
This rule identifies a discrepancy between actual network connections processed by HAProxy and the corresponding entries in its server-side logs. Such inconsistencies can indicate the presence of a malicious backdoor compiled into or injected into the HAProxy process, designed to facilitate C2 traffic while evading standard monitoring by suppressing or omitting logging for those connections.
This rule detects potential unauthenticated access or unauthorized configuration modifications to an AI agent management backend. It monitors HTTP logs for requests targeting configuration endpoints that lack proper authentication, specifically identifying attempts to retrieve sensitive information like Dify or Coze keys, or attempts to inject new agent configurations via unauthenticated POST requests.
Detects unauthenticated execution of sensitive Grafana MCP tools (tools/call, grafana_api_request) by identifying session IDs that lack a corresponding authenticated handshake event. This behavior is indicative of potential session spoofing or abuse of the MCP protocol where an attacker attempts to interact with privileged tools without performing a valid authentication handshake.
Detects anomalous outbound network connections from the Grafana MCP (Model Context Protocol) server process to internal private IP address ranges. The rule compares current network activity against a 14-day baseline to filter out known data sources, identifying potentially unauthorized lateral movement or internal network probing via server-side request forgery (SSRF) vulnerabilities.
Detects the presence of a specific sentinel string 'data.external.telemetry' within Terraform provisioner job logs, which indicates the execution of unauthorized data-collection logic within a compromised or backdoored Terraform module.
Detects the presence of PHP code execution patterns (eval, base64_decode, system, shell_exec) injected into Magento log or report files (var/report/, var/log/system.log), indicative of log poisoning used for web shell staging.
This rule identifies a discrepancy between actual network connections processed by HAProxy and the corresponding entries in its server-side logs. Such inconsistencies can indicate the presence of a malicious backdoor compiled into or injected into the HAProxy process, designed to facilitate C2 traffic while evading standard monitoring by suppressing or omitting logging for those connections.
Detects high-frequency password reset or credential recovery attempts originating from a single source IP against multiple user accounts. This behavior is indicative of an automated enumeration or brute-force attack, potentially exploiting vulnerabilities such as the referenced CVE-2026-18963.
Detects high-frequency password reset or credential recovery attempts originating from a single source IP against multiple user accounts. This behavior is indicative of an automated enumeration or brute-force attack, potentially exploiting vulnerabilities such as the referenced CVE-2026-18963.
This rule detects potentially malicious JavaScript files associated with known malicious patterns or suspicious 'preinstall' script behavior within Node.js environments. It monitors for the execution of specific suspicious filenames or the use of common Node.js package management tools (npm, yarn, etc.) in contexts suggesting code generation or build-time abuse.
Detects anomalous unauthenticated password reset requests targeting the Keycloak reset-credentials endpoint. The rule monitors for high-volume password reset attempts from single source IP addresses, specifically flagging those that target common administrative usernames or attempt to reset passwords across multiple distinct user accounts, which may indicate exploitation attempts (e.g., CVE-2026-18963).
This rule detects instances where a user account performs a password update without any preceding or associated account verification events (such as email verification, reset requests, or action execution) within a short timeframe. This behavior can be indicative of account compromise where an adversary modifies credentials for a compromised user account.
Page 43 of 83