Template catalog

What cxg can actually find. Every check in cert-x-gen-templates, the open-source template corpus, each one with a page setting out what it looks for and a link to the file it is written in.

Browse

Checks live at templates/<category>/<subject>/, one file per check, and the category set is fixed. See The template corpus for why. Each link below opens the table at the bottom of this page with that filter applied.

By category

ai · cli-baseline · databases · devops · messaging · monitoring · network · recon · tooling · web

By severity

critical · high · medium · low · info

By language

c · cpp · go · java · javascript · perl · php · python · ruby · rust · shell · yaml

By target kind

cli · http

A minority of templates declare @target_kinds. It is a convention of the templates repository rather than a field cxg’s parser reads, so a template without one is not restricted to any target. Its header does not say. The Target column below is blank for those.

By weakness class

Every CWE the corpus names, with the checks that name it.

Where these fields come from

Each row is read from the template file itself, never from a summary of it. For the eleven annotation languages that means the @field: header cxg parses out of the first 50 lines; for YAML templates it means the document’s own keys, because the YAML engine deserializes the document and never reads comments. The template schemas reference is the authority for both.

A blank cell means the file does not carry that field. It is not a gap this page filled in from somewhere else.

Templates

SearchCategoryLanguageSeverityTarget

ai

Template	What it checks	Language	Severity	Target	Source
`agent-extension-install-path-containment`	Agent extension installer joins untrusted skill/plugin names and archive members without containing them	`shell`	high	`cli`	file
`agent-plugin-loader-conformance`	Agent-plugin launch client violates Agent Plugins 1.0 loader containment and executes load-time side effects	`shell`	high	`cli`	file
`agent-skill-hidden-instruction-trust`	Agent skill executes a concealed directive with no consent boundary between loaded and executed	`python`	high	`cli`	file
`ai-assisted-fuzzing-sqli-seed-corpus`	AI-Assisted Fuzzing SQLi Seed Corpus	`python`	info		file
`api-pull-access-check`	Unauthorized API Pull Access Check	`yaml`	critical		file
`claude-code-sed-bypass-usage`	Claude Code sed DSL Bypass Indicators (Static Scan)	`python`	medium		file
`coding-agent-command-trace-composition`	Coding-agent command validator approves a command trace whose composition executes a dangerous action	`shell`	high	`cli`	file
`coding-agent-config-allowlist-trust`	Coding-agent CLI takes its command allowlist from attacker-writable configuration, or matches it by command name	`shell`	high	`cli`	file
`coding-agent-cross-tool-config-bleed`	Coding agent honours another agent’s configuration, and one marketplace consent arms more than one agent	`shell`	high	`cli`	file
`coding-agent-deny-rule-reachability`	A coding agent’s strongest deny rule is reachable around - the same protected file is read through syntactically different channels, and through a symlink swapped between check and use	`shell`	high	`cli`	file
`coding-agent-git-config-exec`	Coding-agent CLI lets an untrusted workspace’s own .git/config execute code through its background git calls (GitSpawn)	`shell`	high	`cli`	file
`coding-agent-hook-gate-integrity`	User-installed hook gate does not hold - a pre-execution security hook is bypassed, disarmed by repo-scoped config, or fails open	`shell`	high	`cli`	file
`coding-agent-project-local-config-trust`	Coding-agent CLI executes project-local hooks from a world-writable workspace or a world-writable ancestor of one	`shell`	high	`cli`	file
`coding-agent-repo-config-autoexec`	Coding-agent CLI honours repo-supplied configuration on first open, before any workspace-trust prompt	`shell`	high	`cli`	file
`coding-agent-repo-config-credential-redirect`	Coding-agent CLI applies repo-supplied endpoint environment, sending its own API credential to a host the repository chose	`shell`	critical	`cli`	file
`coding-agent-sandbox-perimeter-enforcement`	Declared agent sandbox is decorative - sandboxEnabled removes the human approval gate but does not confine the process	`shell`	high	`cli`	file
`coding-agent-sandbox-trust-handoff`	Agent sandbox contains the process but not the files it writes - deferred consumers execute agent-authored surfaces after the sandbox exits	`shell`	high	`cli`	file
`coding-agent-shared-config-trust`	Coding-agent CLI honours managed configuration from a world-writable shared path	`shell`	high	`cli`	file
`coding-agent-workspace-interpreter-shadowing`	Coding-agent CLI runs an interpreter in an untrusted workspace whose files shadow the standard library	`shell`	high	`cli`	file
`copilot-yolo-autoapprove-enabled`	GitHub Copilot YOLO Mode Enabled (Static Scan)	`python`	high		file
`cursor-mcpoison-config-risk`	Cursor MCP Config Command Risk (MCPoison)	`python`	high		file
`exposed-ollama-api-version`	Exposed Ollama API Version Endpoint	`yaml`	medium		file
`flowise-custommcp-command-endpoint-exposed`	Flowise CustomMCP Command Endpoint Exposed	`yaml`	critical		file
`flowise-custommcp-js-eval-exposed`	Flowise CustomMCP JS Eval Endpoint Exposed	`yaml`	critical		file
`invokeai-model-install-exposed`	InvokeAI Unauthenticated Model Install Endpoint	`yaml`	critical		file
`mcp-broken-token-validation`	MCP Broken Token Validation (Accepts Forged Tokens)	`python`	high		file
`mcp-cache-scope-identity-leak`	MCP CacheableResult Cross-Identity Leak (identity-dependent response marked publicly cacheable)	`python`	high	`http`	file
`mcp-child-env-inheritance`	MCP host hands a locally-spawned stdio server the operator’s entire environment	`shell`	high	`cli`	file
`mcp-client-header-value-encoding`	MCP Client x-mcp-header Value-Encoding Failure (Model-Controlled HTTP Header Injection)	`python`	high	`cli`	file
`mcp-client-mrtr-conformance`	MCP Client MRTR Conformance (inputRequests Provenance Boundary and the resultType Default)	`python`	high	`cli`	file
`mcp-client-oauth-issuer-binding`	MCP Client OAuth Issuer Binding (Authorization-Server Mix-Up / Cross-Issuer Credential Reuse)	`python`	high	`cli`	file
`mcp-client-untrusted-annotation-approval`	MCP Client Auto-Approves a Tool on an Untrusted Server’s `readOnlyHint` Annotation	`python`	high	`cli`	file
`mcp-credential-exposure`	MCP Credential / Secret Exposure via Resources	`python`	critical	`http`	file
`mcp-excessive-scope-proof`	MCP Out-of-Scope Tool Action (Behavioural Scope Proof)	`python`	high	`http`, `cli`	file
`mcp-excessive-tool-permissions`	MCP Excessive Tool Permissions (Dangerous Capability Exposure)	`python`	high		file
`mcp-handle-binding-integrity`	MCP Handle Binding and requestState Integrity (Stateless Core)	`python`	high	`http`, `cli`	file
`mcp-invisible-unicode-poisoning`	MCP Invisible-Unicode Tool Poisoning (Approval-View Fidelity Gap)	`python`	high	`http`	file
`mcp-manifest-runtime-divergence`	MCP Declared-Manifest / Runtime-Surface Divergence	`python`	high	`http`	file
`mcp-method-desync`	MCP Routing-Header / Body Method Desync (Gateway Authz Bypass)	`python`	high	`http`	file
`mcp-oauth-consent-dcr-abuse`	MCP OAuth Consent-Layer Confused Deputy via Open Dynamic Client Registration	`python`	high	`http`	file
`mcp-rug-pull-detection`	MCP Rug Pull (Silent Tool Definition Mutation)	`python`	high		file
`mcp-token-audience-confusion`	MCP Token Passthrough / Audience Confusion (Confused Deputy)	`python`	high	`http`	file
`mcp-tool-poisoning`	MCP Tool Poisoning (Model-Directed Instructions in Tool Metadata)	`python`	high	`http`	file
`mcp-unauthenticated-access`	Unauthenticated MCP Server Exposure	`python`	high		file
`ml-unsafe-deserialization-usage`	ML Unsafe Deserialization Usage (Static Scan)	`python`	medium		file
`ollama-unauth-api-generate`	Ollama Unauthenticated /api/generate Access	`yaml`	high		file
`pytorch-unsafe-load-usage`	PyTorch Unsafe torch.load Usage (Static Scan)	`python`	high		file
`torchserve-management-api-exposed`	TorchServe Management API Exposed	`yaml`	high		file
`triton-model-control-exposed`	NVIDIA Triton Model Repository Control Exposed	`yaml`	high		file

cli-baseline

Template	What it checks	Language	Severity	Target	Source
`cli-baseline-b01-argument-injection`	CLI Baseline B01 - argument injection via un-separated argv	`shell`	high	`cli`	file
`cli-baseline-b02-command-injection`	CLI Baseline B02 - OS command injection through an argument	`shell`	critical	`cli`	file
`cli-baseline-b03-path-traversal`	CLI Baseline B03 - path traversal in a file-naming argument	`shell`	high	`cli`	file
`cli-baseline-b04-archive-traversal`	CLI Baseline B04 - zip-slip / archive extraction traversal	`shell`	high	`cli`	file
`cli-baseline-b05-argv-secrets`	CLI Baseline B05 - credentials exposed in the process table	`shell`	high	`cli`	file
`cli-baseline-b06-insecure-temp-files`	CLI Baseline B06 - insecure temporary file creation	`shell`	medium	`cli`	file
`cli-baseline-b07-path-hijack`	CLI Baseline B07 - untrusted search path / PATH hijack	`shell`	high	`cli`	file
`cli-baseline-b08-terminal-escape`	CLI Baseline B08 - terminal escape sequence injection	`shell`	medium	`cli`	file
`cli-baseline-b09-environment-trust`	CLI Baseline B09 - unvalidated trust in environment variables	`shell`	medium	`cli`	file
`cli-baseline-b10-config-credential-handling`	CLI Baseline B10 - insecure config and credential file handling	`shell`	high	`cli`	file
`cli-baseline-b11-memory-safety`	CLI Baseline B11 - memory-safety defect (instrumentation-dependent)	`shell`	critical	`cli`	file
`cli-baseline-b12-crash-hang`	CLI Baseline B12 - crash or hang on malformed input	`shell`	medium	`cli`	file
`cli-baseline-b13-toctou-symlink-race`	CLI Baseline B13 - TOCTOU / symlink race	`shell`	high	`cli`	file
`cli-baseline-b14-format-string`	CLI Baseline B14 - format-string defect	`shell`	high	`cli`	file

databases

Template	What it checks	Language	Severity	Source
`clickhouse-auth-bypass`	ClickHouse Authentication Bypass	`python`	high	file
`cockroachdb-unauthenticated-access`	CockroachDB Unauthenticated Access Detection	`yaml`	critical	file
`couchdb-default-credentials`	CouchDB Default Credentials Detection	`rust`	critical	file
`elasticsearch-data-exposure`	Elasticsearch Cluster Data Exposure	`python`	critical	file
`elasticsearch-query-injection`	Elasticsearch Query DSL Injection Detector	`python`	high	file
`elasticsearch-unauthenticated`	Elasticsearch Unauthenticated Access Detection	`rust`	critical	file
`influxdb-health-exposed`	InfluxDB Health Endpoint Exposed	`yaml`	medium	file
`memcached-unauthenticated-access`	Memcached Unauthenticated Access Detection	`yaml`	critical	file
`mongodb-injection-deep`	MongoDB NoSQL Injection Deep Analysis	`python`	high	file
`mongodb-unauthenticated`	MongoDB Unauthenticated Access Detection	`python`	critical	file
`mysql-default-credentials`	MySQL Default/Weak Credentials Detection	`python`	critical	file
`postgresql-default-credentials`	PostgreSQL Default Credentials Detection	`go`	critical	file
`postgresql-extension-rce`	postgresql extension rce	`python`	medium	file
`redis-cluster-takeover`	Redis Cluster Takeover Detection	`go`	critical	file
`redis-unauthenticated-access`	Redis Unauthenticated Access Detection	`yaml`	critical	file
`redis-unauthenticated-c`	Redis Unauthenticated Access Detection (C)	`c`	critical	file
`redis-unauthenticated-cpp`	Redis Unauthenticated Access Detection (C++)	`cpp`	critical	file
`redis-unauthenticated-go`	Redis Unauthenticated Access Detection (Go)	`go`	critical	file
`redis-unauthenticated-java`	Redis Unauthenticated Access Detection (Java)	`java`	critical	file
`redis-unauthenticated-javascript`	Redis Unauthenticated Access Detection (JavaScript)	`javascript`	critical	file
`redis-unauthenticated-perl`	Redis Unauthenticated Access Detection (Perl)	`perl`	critical	file
`redis-unauthenticated-php`	Redis Unauthenticated Access Detection (PHP)	`php`	critical	file
`redis-unauthenticated-python`	Redis Unauthenticated Access Detection (Python)	`python`	critical	file
`redis-unauthenticated-ruby`	Redis Unauthenticated Access Detection (Ruby)	`ruby`	critical	file
`redis-unauthenticated-rust`	Redis Unauthenticated Access Detection (Rust)	`rust`	critical	file
`redis-unauthenticated-shell`	Redis Unauthenticated Access Detection (Shell)	`shell`	critical	file

devops

Template	What it checks	Language	Severity	Source
`actions-injection-scanner`	GitHub Actions Injection Scanner	`python`	high	file
`ci-variable-exposure`	CI/CD Variable Exposure Detection	`python`	critical	file
`docker-api-unauth`	Docker API Unauthenticated Access	`go`	critical	file
`docker-registry-unauthenticated`	Docker Registry Unauthenticated Access Detection	`shell`	high	file
`etcd-auth-check`	Etcd Authentication Check	`yaml`	info	file
`etcd-instead-http-request`	Etcd Member List Security Check	`yaml`	medium	file
`etcd-running-unauthenticated-port`	Unauthenticated etcd Service Detection	`shell`	critical	file
`etcd-think-path-wrong`	Etcd Unauthenticated Access	`yaml`	medium	file
`etcd-unauth`	Etcd Unauthenticated Access - Members List	`yaml`	high	file
`etcd-unauthenticated`	Etcd Unauthenticated Member List Detection	`shell`	high	file
`etcd-unauthenticated-access`	Etcd Unauthenticated Access - Version Endpoint	`yaml`	medium	file
`ghes-version-fingerprint`	GitHub Enterprise Server Version Fingerprint	`python`	info	file
`git-history-secret-scan`	Git History Secret Scan	`go`	critical	file
`gitlab-version-fingerprint`	GitLab Version Fingerprint	`python`	info	file
`helm-chart-secrets-leak`	Helm Chart Secrets Leak Detection	`python`	high	file
`istio-pilot-misconfiguration`	Istio Pilot Misconfiguration Detection	`go`	high	file
`jenkins-unauth-rce`	Jenkins Unauthenticated Script Console RCE	`go`	critical	file
`jupyter-unauth-rce`	Jupyter Notebook Unauthenticated RCE	`python`	critical	file
`k8s-etcd-exposed`	Kubernetes etcd Secrets Exposure	`go`	critical	file
`k8s-rbac-misconfiguration`	Kubernetes RBAC Misconfiguration Detection	`go`	critical	file
`kubelet-api-exposure`	Kubelet API Exposure Detection	`go`	critical	file
`kubernetes-api-unauthenticated`	Kubernetes API Unauthenticated Access	`yaml`	critical	file
`pwn-request-scanner`	GitHub Actions Pwn Request Scanner	`python`	critical	file
`runner-token-detection`	runner token detection	`go`	medium	file
`saml-sso-bypass-gitlab`	saml sso bypass gitlab	`python`	medium	file
`service-account-token-abuse`	Kubernetes Service Account Token Abuse	`go`	critical	file

messaging

Template	What it checks	Language	Severity	Source
`kafka-unauthenticated`	Apache Kafka Unauthenticated Access Detection	`shell`	high	file
`kafka-unauthenticated-access`	Kafka Unauthenticated Access Detection	`python`	critical	file
`mqtt-unauthenticated`	MQTT Unauthenticated Access (CONNACK Accepted)	`yaml`	high	file
`nats-unauthenticated-banner`	NATS Banner Indicates No Authentication	`yaml`	medium	file
`rabbitmq-default-credentials`	RabbitMQ Default Credentials Detection	`python`	critical	file
`rabbitmq-management-exposed`	RabbitMQ Management UI Exposed	`yaml`	medium	file
`zookeeper-unauthenticated-access`	Apache Zookeeper Unauthenticated Access	`yaml`	critical	file

monitoring

Template	What it checks	Language	Severity	Source
`cadvisor-exposed-javascript`	cAdvisor Exposed Without Authentication	`javascript`	critical	file
`cadvisor-exposed-python`	cAdvisor Metrics Exposed	`python`	medium	file
`check-metrics-endpoint`	Check /metrics Endpoint Exposure	`yaml`	medium	file
`kibana-api-status-exposed`	Kibana /api/status Exposed	`yaml`	medium	file
`mysql-exporter-exposed`	MySQL Exporter Metrics Exposed	`python`	medium	file
`node-exporter-exposed-javascript`	Node Exporter Exposed Without Authentication	`javascript`	high	file
`node-exporter-exposed-python`	Node Exporter Metrics Exposed	`python`	medium	file
`postgresql-exporter-exposed`	PostgreSQL Exporter Metrics Exposed	`python`	medium	file
`prometheus-node-exporter-exposed`	Prometheus Node Exporter Exposed Without Authentication	`yaml`	high	file
`prometheus-server-exposed-https`	Prometheus Server Exposed (HTTPS)	`python`	high	file
`prometheus-server-exposed-javascript`	Prometheus Server Exposed Without Authentication	`javascript`	high	file
`prometheus-server-exposed-python`	Prometheus Server Exposed	`python`	high	file
`redis-exporter-exposed`	Redis Exporter Exposed Detection	`yaml`	medium	file
`redis-exporter-exposed-python`	Redis Exporter Metrics Exposed	`python`	high	file
`splunk-web-login-exposed`	Splunk Web Login Exposed	`yaml`	medium	file
`splunkd-server-info-exposed`	Splunkd Server Info Exposed	`yaml`	high	file

network

Template	What it checks	Language	Severity	Source
`adb-exposed`	ADB Over TCP Exposed	`yaml`	high	file
`dhcpv6-solicit-response`	DHCPv6 Solicit Response Probe	`python`	info	file
`dns-rebinding-attack`	DNS Rebinding Attack Detection	`go`	critical	file
`dns-udp-service-probe`	DNS UDP Service Probe	`python`	info	file
`dns-zone-transfer`	DNS Zone Transfer (AXFR) Detection	`python`	high	file
`echo-service-exposed`	Echo Service Exposed	`yaml`	low	file
`epmd-node-list-exposed`	EPMD Node List Exposed	`yaml`	medium	file
`finger-service-exposed`	Finger Service Exposed	`yaml`	low	file
`ftp-anonymous-access`	FTP Anonymous Access Detection	`python`	high	file
`grpc-reflection-abuse`	gRPC Reflection API Exposure Detection	`go`	high	file
`http-service-responding`	HTTP Service Responding	`yaml`	info	file
`icmp-echo-reachable`	ICMP Echo Reachability	`python`	info	file
`ident-exposed`	Ident Service Exposed	`yaml`	low	file
`mdns-service-discovery-probe`	mDNS Service Discovery Probe	`python`	info	file
`nbns-name-query-probe`	NBNS Name Query Probe	`python`	info	file
`ndmp-service-exposed`	NDMP Service Exposed	`yaml`	medium	file
`ntp-udp-service-probe`	NTP UDP Service Probe	`python`	info	file
`port-scanner-async`	High-Speed Async TCP Port Scanner	`rust`	info	file
`rmi-service-enumeration`	RMI Service Enumeration	`java`	high	file
`rsync-banner-exposed`	Rsync Daemon Banner Exposed	`yaml`	medium	file
`smart-install-exposed`	Cisco Smart Install Exposed	`yaml`	high	file
`smtp-open-relay`	SMTP Open Relay Detection	`python`	high	file
`snmp-default-community`	SNMP Default Community String Detection	`shell`	high	file
`socks5-no-auth`	SOCKS5 No-Authentication Proxy	`yaml`	high	file
`ssdp-msearch-response`	SSDP M-SEARCH Response Probe	`python`	info	file
`tacacs-service-exposed`	TACACS+ Service Exposed	`yaml`	medium	file
`tcp-banner-probe`	TCP Banner Probe	`python`	info	file
`tcp-port-reachability`	TCP Port Reachability	`python`	info	file
`tftp-service-exposed`	TFTP Service Exposed	`yaml`	medium	file
`tls-certificate-deep-analysis`	TLS Certificate Deep Analysis	`rust`	high	file
`vnc-no-auth`	VNC No Authentication Detection	`c`	critical	file
`websocket-message-fuzzer`	websocket message fuzzer	`javascript`	medium	file
`whois-service-exposed`	WHOIS Service Exposed	`yaml`	low	file
`wsd-probe-response`	WSD Probe Response	`python`	info	file

recon

Template	What it checks	Language	Severity	Target	Source
`system-context-recon`	System Context Reconnaissance	`shell`	info		file

tooling

Template	What it checks	Language	Severity	Target	Source
`supply-chain-install-credential-access`	Supply chain - install-time credential access and workflow planting	`shell`	critical	`cli`	file
`supply-chain-install-hook-behavior`	Supply chain - install and import-time hook behaviour	`shell`	critical	`cli`	file

web

Template	What it checks	Language	Severity	Source
`auth-bypass-flow`	Authentication Bypass with Session Flow	`yaml`	critical	file
`auth-token-logout-reuse`	Authentication Token Reuse After Logout	`python`	high	file
`deserialization-gadget-scan`	Deserialization Gadget Scan	`java`	critical	file
`directory-listing-common-paths`	Directory Listing Exposure (Common Paths)	`yaml`	medium	file
`directory-traversal`	Directory Traversal Detection	`c`	high	file
`example-http-check`	Example HTTP Security Check	`yaml`	info	file
`forced-browse-auth-bypass`	Forced Browse Auth Bypass — Protected Endpoints Without Auth	`python`	high	file
`graphql-user-enumeration`	GraphQL User Enumeration (CVE-2021-4191)	`python`	medium	file
`http-header-injection`	HTTP Header Injection Detection	`python`	medium	file
`http-service-detection`	HTTP Service Detection	`yaml`	info	file
`http2-rapid-reset`	HTTP/2 Rapid Reset Detection (CVE-2023-44487)	`go`	critical	file
`idor-bola-horizontal`	IDOR / BOLA Horizontal Privilege Escalation	`python`	high	file
`jwt-alg-confusion`	JWT Algorithm Confusion (RS256 → HS256 / none)	`python`	critical	file
`jwt-role-tampering`	JWT Role / Privilege Claim Tampering	`python`	critical	file
`log4shell-detection`	Log4Shell (CVE-2021-44228) Vulnerability Detection	`shell`	critical	file
`mass-assignment-update`	Mass Assignment via Profile / Object Update Endpoint	`python`	high	file
`password-reset-enum`	Password Reset User Enumeration via Response Differences	`python`	medium	file
`password-reset-takeover`	password reset takeover	`python`	medium	file
`prototype-pollution`	prototype pollution	`javascript`	medium	file
`race-condition-exploit`	Race Condition Detection (TOCTOU)	`go`	critical	file
`rate-limit-auth-bypass`	Missing or Bypassable Rate Limit on Authentication Endpoint	`python`	high	file
`response-integrity-audit`	Response Integrity Audit (Cache Poisoning & HPP)	`yaml`	medium	file
`sensitive-data-exposure`	Sensitive Data Exposure Detection	`yaml`	medium	file
`server-side-js-injection`	server side js injection	`javascript`	medium	file
`session-fixation-check`	Session Fixation — Token Unchanged After Login	`python`	high	file
`spring4shell-detection`	Spring4Shell Detection (CVE-2022-22965)	`java`	critical	file
`sql-injection-basic`	SQL Injection Detection (Basic)	`yaml`	critical	file
`sql-injection-detection-c`	SQL Injection Detection	`c`	critical	file
`ssti-engine-fingerprint`	Server-Side Template Injection Engine Fingerprint	`python`	high	file
`timing-attack-detection`	Blind SQL Injection (Time-Based)	`yaml`	high	file
`vertical-privesc-admin`	Vertical Privilege Escalation to Admin Endpoints	`python`	critical	file
`xss-detection-c`	Cross-Site Scripting (XSS) Detection	`c`	high	file