SecurityDetections
Detection Catalogue
Filesystem
Commands that recursively or broadly delete files or directories.
Rules:
FS001criticalRecursive home directory deletion
Detects rm -rf or equivalent targeting home directory or root filesystem
FS002highRecursive directory deletion
Detects recursive deletion of directories
FS003warnWildcard deletion in sensitive directory
Detects deletion with wildcards in potentially sensitive locations
Database
SQL and migration commands that drop, truncate or wipe tables and data.
Rules:
DB001criticalDROP TABLE or DATABASE
Detects SQL DROP TABLE or DROP DATABASE commands
DB002criticalDELETE without WHERE
Detects DELETE statements without WHERE clause
DB003highTRUNCATE TABLE
Detects TRUNCATE TABLE commands
DB004highDangerous migration reset
Detects migration commands that reset or drop all tables (prisma push --force-reset, drizzle-kit drop)
Cloud infrastructure
Commands that delete cloud or Kubernetes resources.
Rules:
CLOUD001criticalTerraform destroy
Detects terraform destroy commands
CLOUD002criticalAWS resource deletion
Detects AWS CLI commands that delete resources (ec2 terminate, s3 rb, rds delete, etc.)
CLOUD003criticalGCP resource deletion
Detects gcloud commands that delete compute, storage, or database resources
CLOUD004criticalAzure resource deletion
Detects az commands that delete resources
CLOUD005criticalKubernetes resource deletion
Detects kubectl delete commands on production-like resources
Production targeting
Database connections and cloud or Kubernetes commands aimed at production.
Rules:
PROD001highProduction database connection
Detects database connections to production hosts (non-localhost, containing 'prod')
PROD002highProduction cloud context
Detects cloud commands targeting production accounts or regions
PROD003highProduction Kubernetes context
Detects kubectl commands with production context names
Secrets
Credential files that are read, staged or sent over the network.
Rules:
SECRET001highSecret file access followed by network egress
Detects reading of credential files followed by network operations within a short window
SECRET002highCredential file staging
Detects copying or staging of credential files to /tmp or other staging locations
SECRET003highSecret exfiltration via network tools
Detects curl/wget/scp/ssh uploading or sending secret files or secret-like values to remote hosts
Git
Force pushes, hard resets and branch deletion.
Rules:
GIT001highForce push to main branch
Detects git push --force to main, master, or production branches
GIT002warnHard reset of branch
Detects git reset --hard which discards uncommitted changes
GIT003warnBranch deletion
Detects deletion of git branches
Safety checks
Flags and settings that bypass permission prompts, git hooks, tests or file permissions.
Rules:
SAFETY001highAI CLI permission bypass flags
Detects AI CLI tools invoked with permission bypass flags (--dangerously-skip-permissions, --yolo, --trust-all-tools)
SAFETY002warnGit hooks bypass
Detects git operations with --no-verify flag
SAFETY003warnTest skip or disable
Detects commands that skip tests or set test environment variables to disable test suites
SAFETY004highDangerous file permissions
Detects chmod 777 or equivalent overly permissive settings
Risky installs
Piped remote scripts and package names that look like typosquats.
Rules:
INSTALL001highCurl pipe to shell
Detects curl|sh, wget|bash, and similar piped execution patterns
INSTALL002infoSuspicious package name
Detects npm/pip/gem install of packages with typosquat-like patterns (multiple hyphens, l33tspeak, common typos)
Prompt injection
Instructions in fetched content or tool output that try to steer the agent.
Rules:
INJECT001highPrompt injection in fetched content
Detects fetched content (web, issue, file) containing agent-override instructions
INJECT002warnHidden instructions in tool output
Detects tool output containing instructions formatted to look like system messages
Multi-step
Several events in one session that together match an attack pattern.
Rules:
MULTI001criticalInjection followed by sensitive read and network egress
Detects three-step sequence: injection content, then sensitive file access, then network upload