Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
1,084 detections
Filters
Last updated
All Time
Detection languages
418
193
163
156
81
Contributors
381
251
92
71
40
Categories
498
271
255
135
123
Platforms
443
441
267
121
114
Products / Services
10,665
9,773
6,704
4,421
3,749
1,084
MITRE Techniques
278
229
180
177
151
CVEs
17
15
13
13
8
IDS Classtypes
4
IDS Protocols
4
Detects instances where Zscaler AI Protect (AI Guard) flags attempts to perform prompt injection or jailbreak against Large Language Model (LLM) or GenAI applications. The rule monitors blocked requests and allowed requests that exceed a high confidence score threshold (0.75), identifying potential malicious interactions with generative AI services.
This rule detects potential brute force or password spraying activity by identifying a user account that experiences at least 5 login failures followed by a successful login for the same application and source IP address within a one-hour window. This pattern is indicative of an attacker who has successfully guessed credentials after multiple attempts.
Detects an applicant identity submitting an abnormally high volume of job applications across many companies within a rolling window, corroborated by resume or portfolio content reused across multiple applicant identities — consistent with PurpleDelta fabricated-persona application tempo.
Detects instances where Zscaler AI Protect (AI Guard) flags attempts to perform prompt injection or jailbreak against Large Language Model (LLM) or GenAI applications. The rule monitors blocked requests and allowed requests that exceed a high confidence score threshold (0.75), identifying potential malicious interactions with generative AI services.
Detects high-volume file deletion or modification activity within sanctioned cloud storage applications, monitored via Netskope. This behavior is indicative of potential ransomware, malicious bulk data destruction, or an insider threat attempting to sabotage data.
Detects instances where an administrator account experiences multiple authentication failures within a 15-minute window, followed by a successful authentication attempt from the same client IP address. This pattern is indicative of potential brute force or password spraying activity where the attacker successfully guessed the credentials after multiple failed attempts.
KQL Query
Detects instances where an administrator account experiences multiple authentication failures within a 15-minute window, followed by a successful authentication attempt from the same client IP address. This pattern is indicative of potential brute force or password spraying activity where the attacker successfully guessed the credentials after multiple failed attempts.
Detects potential SQL injection attempts targeting OGC-compliant web services (such as GeoServer) via specific query parameters like 'cql_filter' or 'filter'. The rule identifies malicious patterns including common SQL keywords and syntax (union, select, information_schema, pg_sleep) within the request URI, query string, or body.
This rule detects potential brute force or password spraying activity by identifying a user account that experiences at least 5 login failures followed by a successful login for the same application and source IP address within a one-hour window. This pattern is indicative of an attacker who has successfully guessed credentials after multiple attempts.
Detects the execution of administrative or system-modifying tools (e.g., bash, sh, chroot, mount) or suspicious command lines involving root-level path manipulation or capability inspection (capsh) initiated by office applications, sandbox host processes, or specialized Excel engines. This behavior is indicative of a potential container breakout, privilege escalation, or post-exploitation activity originating from an initial access vector.
KQL Query
Detects devices exhibiting sustained high CPU or memory utilization (>= 90%) for a duration of at least 30 minutes, which may indicate resource exhaustion, performance degradation, or potentially malicious activity such as cryptocurrency mining, denial-of-service, or other unauthorized resource-intensive processes.
This rule monitors SCIM (System for Cross-domain Identity Management) logs for high-volume identity management operations, including account creation, deletion, or entitlement grants within a short 15-minute window. A trigger occurs if an actor performs 10 or more creations, 10 or more deletions, 10 or more entitlement grants, or 3 or more administrative role grants. This behavior may indicate an account compromise performing bulk manipulation of identities to maintain persistence or elevate privileges.
Detects the execution of administrative or system-modifying tools (e.g., bash, sh, chroot, mount) or suspicious command lines involving root-level path manipulation or capability inspection (capsh) initiated by office applications, sandbox host processes, or specialized Excel engines. This behavior is indicative of a potential container breakout, privilege escalation, or post-exploitation activity originating from an initial access vector.
Detects high-volume file deletion or modification activity within sanctioned cloud storage applications, monitored via Netskope. This behavior is indicative of potential ransomware, malicious bulk data destruction, or an insider threat attempting to sabotage data.
Detects execution of a 'setup.mjs' preinstall lifecycle hook, the stage-1 loader used in the
4 August 2026 compromise of the keyv and cacheable npm namespaces. The trojanised package.json
adds "preinstall": "node setup.mjs" while leaving dist/ byte-identical to the last clean
release, so the only execution-time signal is the hook itself. The loader downloads a
standalone Bun runtime and executes an obfuscated second stage (Math_Symbol.js).
4 August 2026 compromise of the keyv and cacheable npm namespaces. The trojanised package.json
adds "preinstall": "node setup.mjs" while leaving dist/ byte-identical to the last clean
release, so the only execution-time signal is the hook itself. The loader downloads a
standalone Bun runtime and executes an obfuscated second stage (Math_Symbol.js).
This rule monitors SCIM (System for Cross-domain Identity Management) logs for high-volume identity management operations, including account creation, deletion, or entitlement grants within a short 15-minute window. A trigger occurs if an actor performs 10 or more creations, 10 or more deletions, 10 or more entitlement grants, or 3 or more administrative role grants. This behavior may indicate an account compromise performing bulk manipulation of identities to maintain persistence or elevate privileges.
Imperva Site WAF Protection Disabled
Cortex XDR
Detects an audit event in Imperva Cloud WAF where a site's protection state is changed to disabled. Disabling WAF protection reduces the security posture of the protected application and may indicate an attempt to impair defensive controls.
This rule detects the creation of diagnostic crash-dump files within known engineering log directories (e.g., slog, engineermode, vendor/log) within 30 minutes of EngineerModeActivity or YLog execution. This pattern is indicative of potential forensic evidence from an induced modem crash, consistent with known exploit chains targeting specific mobile device vulnerabilities.
Detects a Node or Bun process reaching the cloud instance metadata service. The keyv/cacheable
stage-2 collector targets EC2 IMDS and ECS task metadata as the entry point to a broader sweep
covering AWS credential chains and Secrets Manager across all regions, GCP service account
keys, Azure client secrets, HashiCorp Vault tokens, Kubernetes service account tokens, GitHub
Actions OIDC request tokens and npm tokens. On a build agent this pattern is the clearest
signal that credential theft is in progress.
stage-2 collector targets EC2 IMDS and ECS task metadata as the entry point to a broader sweep
covering AWS credential chains and Secrets Manager across all regions, GCP service account
keys, Azure client secrets, HashiCorp Vault tokens, Kubernetes service account tokens, GitHub
Actions OIDC request tokens and npm tokens. On a build agent this pattern is the clearest
signal that credential theft is in progress.
Page 33 of 52