nuclei/SYNTAX-REFERENCE.md

39 KiB
Executable File

Template

Template is a YAML input file which defines all the requests and other metadata for a template.


id string

ID is the unique id for the template.

Good IDs

A good ID uniquely identifies what the requests in the template are doing. Let's say you have a template that identifies a git-config file on the webservers, a good name would be git-config-exposure. Another example name is azure-apps-nxdomain-takeover.

Examples:

# ID Example
id: CVE-2021-19520

info model.Info

Info contains metadata information about the template.

Examples:

info:
    name: Argument Injection in Ruby Dragonfly
    author: 0xspara
    tags: cve,cve2021,rce,ruby
    reference: https://zxsecurity.co.nz/research/argunment-injection-ruby-dragonfly/
    severity: high

requests []http.Request

Requests contains the http request to make in the template.

Examples:

requests:
    matchers:
        - type: word
          words:
            - '[core]'
        - type: dsl
          condition: and
          dsl:
            - '!contains(tolower(body), ''<html'')'
            - '!contains(tolower(body), ''<body'')'
        - type: status
          status:
            - 200
    matchers-condition: and
    path:
        - '{{BaseURL}}/.git/config'
    method: GET

dns []dns.Request

DNS contains the dns request to make in the template

Examples:

dns:
    extractors:
        - type: regex
          regex:
            - ec2-[-\d]+\.compute[-\d]*\.amazonaws\.com
            - ec2-[-\d]+\.[\w\d\-]+\.compute[-\d]*\.amazonaws\.com
    name: '{{FQDN}}'
    type: CNAME
    class: inet
    retries: 2
    recursion: true

file []file.Request

File contains the file request to make in the template

Examples:

file:
    extractors:
        - type: regex
          regex:
            - amzn\.mws\.[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}
    extensions:
        - all

network []network.Request

Network contains the network request to make in the template

Examples:

network:
    host:
        - '{{Hostname}}'
        - '{{Hostname}}:2181'
    inputs:
        - data: "envi\r\nquit\r\n"
    read-size: 2048
    matchers:
        - type: word
          words:
            - zookeeper.version

headless []headless.Request

Headless contains the headless request to make in the template.


Workflows is a list of workflows to execute for a template.


self-contained bool

Self Contained marks Requests for the template as self-contained


model.Info

Info contains metadata information about a template

Appears in:

name: Argument Injection in Ruby Dragonfly
author: 0xspara
tags: cve,cve2021,rce,ruby
reference: https://zxsecurity.co.nz/research/argunment-injection-ruby-dragonfly/
severity: high

name string

Name should be good short summary that identifies what the template does.

Examples:

name: bower.json file disclosure
name: Nagios Default Credentials Check

Author of the template.

Multiple values can also be specified separated by commas.

Examples:

author: <username>

Any tags for the template.

Multiple values can also be specified separated by commas.

Examples:

# Example tags
tags: cve,cve2019,grafana,auth-bypass,dos

description string

Description of the template.

You can go in-depth here on what the template actually does.

Examples:

description: Bower is a package manager which stores package information in the bower.json file
description: Subversion ALM for the enterprise before 8.8.2 allows reflected XSS at multiple locations

References for the template.

This should contain links relevant to the template.

Examples:

reference:
    - https://github.com/strapi/strapi
    - https://github.com/getgrav/grav

severity severity.Holder

Severity of the template.

Valid values:

  • info

  • low

  • medium

  • high

  • critical


metadata map[string]string

Metadata of the template.

Examples:

metadata:
    customField1: customValue1

classification model.Classification

Classification contains classification information about the template.


remediation string

Remediation steps for the template.

You can go in-depth here on how to mitigate the problem found by this template.

Examples:

remediation: Change the default administrative username and password of Apache ActiveMQ by editing the file jetty-realm.properties

stringslice.StringSlice

StringSlice represents a single (in-lined) or multiple string value(s). The unmarshaller does not automatically convert in-lined strings to []string, hence the interface{} type is required.

Appears in:

<username>
# Example tags
cve,cve2019,grafana,auth-bypass,dos
- https://github.com/strapi/strapi
- https://github.com/getgrav/grav
CVE-2020-14420
CWE-22

severity.Holder

Holder holds a Severity type. Required for un/marshalling purposes

Appears in:

model.Classification

Appears in:


CVE ID for the template

Examples:

cve-id: CVE-2020-14420

CWE ID for the template.

Examples:

cwe-id: CWE-22

cvss-metrics string

CVSS Metrics for the template.

Examples:

cvss-metrics: 3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

cvss-score float64

CVSS Score for the template.

Examples:

cvss-score: "9.8"

http.Request

Request contains a http request to be made from a template

Appears in:

matchers:
    - type: word
      words:
        - '[core]'
    - type: dsl
      condition: and
      dsl:
        - '!contains(tolower(body), ''<html'')'
        - '!contains(tolower(body), ''<body'')'
    - type: status
      status:
        - 200
matchers-condition: and
path:
    - '{{BaseURL}}/.git/config'
method: GET

matchers []matchers.Matcher

Matchers contains the detection mechanism for the request to identify whether the request was successful by doing pattern matching on request/responses.

Multiple matchers can be combined with matcher-condition flag which accepts either and or or as argument.


extractors []extractors.Extractor

Extractors contains the extraction mechanism for the request to identify and extract parts of the response.


matchers-condition string

MatchersCondition is the condition between the matchers. Default is OR.

Valid values:

  • and

  • or


path []string

Path contains the path/s for the HTTP requests. It supports variables as placeholders.

Examples:

# Some example path values
path:
    - '{{BaseURL}}'
    - '{{BaseURL}}/+CSCOU+/../+CSCOE+/files/file_list.json?path=/sessions'

raw []string

Raw contains HTTP Requests in Raw format.

Examples:

# Some example raw requests
raw:
    - |-
      GET /etc/passwd HTTP/1.1
      Host:
      Content-Length: 4      
    - |-
      POST /.%0d./.%0d./.%0d./.%0d./bin/sh HTTP/1.1
      Host: {{Hostname}}
      User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:71.0) Gecko/20100101 Firefox/71.0
      Content-Length: 1
      Connection: close

      echo
      echo
      cat /etc/passwd 2>&1      

id string

ID is the optional id of the request


name string

Name is the optional name of the request.

If a name is specified, all the named request in a template can be matched upon in a combined manner allowing multirequest based matchers.


attack string

Attack is the type of payload combinations to perform.

batteringram is same payload into all of the defined payload positions at once, pitchfork combines multiple payload sets and clusterbomb generates permutations and combinations for all payloads.

Valid values:

  • batteringram

  • pitchfork

  • clusterbomb


method string

Method is the HTTP Request Method.

Valid values:

  • GET

  • HEAD

  • POST

  • PUT

  • DELETE

  • CONNECT

  • OPTIONS

  • TRACE

  • PATCH

  • PURGE


body string

Body is an optional parameter which contains HTTP Request body.

Examples:

# Same Body for a Login POST request
body: username=test&password=test

payloads map[string]interface{}

Payloads contains any payloads for the current request.

Payloads support both key-values combinations where a list of payloads is provided, or optionally a single file can also be provided as payload which will be read on run-time.


headers map[string]string

Headers contains HTTP Headers to send with the request.

Examples:

headers:
    Any-Header: Any-Value
    Content-Length: "1"
    Content-Type: application/x-www-form-urlencoded

race_count int

RaceCount is the number of times to send a request in Race Condition Attack.

Examples:

# Send a request 5 times
race_count: 5

max-redirects int

MaxRedirects is the maximum number of redirects that should be followed.

Examples:

# Follow up to 5 redirects
max-redirects: 5

pipeline-concurrent-connections int

PipelineConcurrentConnections is number of connections to create during pipelining.

Examples:

# Create 40 concurrent connections
pipeline-concurrent-connections: 40

pipeline-requests-per-connection int

PipelineRequestsPerConnection is number of requests to send per connection when pipelining.

Examples:

# Send 100 requests per pipeline connection
pipeline-requests-per-connection: 100

threads int

Threads specifies number of threads to use sending requests. This enables Connection Pooling.

Connection: Close attribute must not be used in request while using threads flag, otherwise pooling will fail and engine will continue to close connections after requests.

Examples:

# Send requests using 10 concurrent threads
threads: 10

max-size int

MaxSize is the maximum size of http response body to read in bytes.

Examples:

# Read max 2048 bytes of the response
max-size: 2048

cookie-reuse bool

CookieReuse is an optional setting that enables cookie reuse for all requests defined in raw section.


redirects bool

Redirects specifies whether redirects should be followed by the HTTP Client.

This can be used in conjunction with max-redirects to control the HTTP request redirects.


pipeline bool

Pipeline defines if the attack should be performed with HTTP 1.1 Pipelining

All requests must be idempotent (GET/POST). This can be used for race conditions/billions requests.


unsafe bool

Unsafe specifies whether to use rawhttp engine for sending Non RFC-Compliant requests.

This uses the rawhttp engine to achieve complete control over the request, with no normalization performed by the client.


race bool

Race determines if all the request have to be attempted at the same time (Race Condition)

The actual number of requests that will be sent is determined by the race_count field.


req-condition bool

ReqCondition automatically assigns numbers to requests and preserves their history.

This allows matching on them later for multi-request conditions.


stop-at-first-match bool

StopAtFirstMatch stops the execution of the requests and template as soon as a match is found.


skip-variables-check bool

SkipVariablesCheck skips the check for unresolved variables in request


matchers.Matcher

Matcher is used to match a part in the output from a protocol.

Appears in:


type string

Type is the type of the matcher.

Valid values:

  • status

  • size

  • word

  • regex

  • binary

  • dsl


condition string

Condition is the optional condition between two matcher variables. By default, the condition is assumed to be OR.

Valid values:

  • and

  • or


part string

Part is the part of the request response to match data from.

Each protocol exposes a lot of different parts which are well documented in docs for each request type.

Examples:

part: body
part: raw

negative bool

Negative specifies if the match should be reversed It will only match if the condition is not true.


name string

Name of the matcher. Name should be lowercase and must not contain spaces or underscores (_).

Examples:

name: cookie-matcher

status []int

Status are the acceptable status codes for the response.

Examples:

status:
    - 200
    - 302

size []int

Size is the acceptable size for the response

Examples:

size:
    - 3029
    - 2042

words []string

Words contains word patterns required to be present in the response part.

Examples:

# Match for outlook mail protection domain
words:
    - mail.protection.outlook.com
# Match for application/json in response headers
words:
    - application/json

regex []string

Regex contains Regular Expression patterns required to be present in the response part.

Examples:

# Match for Linkerd Service via Regex
regex:
    - (?mi)^Via\\s*?:.*?linkerd.*$
# Match for Open Redirect via Location header
regex:
    - (?m)^(?:Location\\s*?:\\s*?)(?:https?://|//)?(?:[a-zA-Z0-9\\-_\\.@]*)example\\.com.*$

binary []string

Binary are the binary patterns required to be present in the response part.

Examples:

# Match for Springboot Heapdump Actuator "JAVA PROFILE", "HPROF", "Gunzip magic byte"
binary:
    - 4a4156412050524f46494c45
    - 4850524f46
    - 1f8b080000000000
# Match for 7zip files
binary:
    - 377ABCAF271C

dsl []string

DSL are the dsl expressions that will be evaluated as part of nuclei matching rules. A list of these helper functions are available here.

Examples:

# DSL Matcher for package.json file
dsl:
    - contains(body, 'packages') && contains(tolower(all_headers), 'application/octet-stream') && status_code == 200
# DSL Matcher for missing strict transport security header
dsl:
    - '!contains(tolower(all_headers), ''''strict-transport-security'''')'

encoding string

Encoding specifies the encoding for the words field if any.

Valid values:

  • hex

extractors.Extractor

Extractor is used to extract part of response using a regex.

Appears in:


name string

Name of the extractor. Name should be lowercase and must not contain spaces or underscores (_).

Examples:

name: cookie-extractor

type string

Type is the type of the extractor.

Valid values:

  • regex

  • kval

  • json

  • xpath


regex []string

Regex contains the regular expression patterns to extract from a part.

Go regex engine does not support lookaheads or lookbehinds, so as a result they are also not supported in nuclei.

Examples:

# Braintree Access Token Regex
regex:
    - access_token\$production\$[0-9a-z]{16}\$[0-9a-f]{32}
# Wordpress Author Extraction regex
regex:
    - Author:(?:[A-Za-z0-9 -\_="]+)?<span(?:[A-Za-z0-9 -\_="]+)?>([A-Za-z0-9]+)<\/span>

group int

Group specifies a numbered group to extract from the regex.

Examples:

# Example Regex Group
group: 1

kval []string

description: | kval contains the key-value pairs present in the HTTP response header. kval extractor can be used to extract HTTP response header and cookie key-value pairs. kval extractor inputs are case-insensitive, and does not support dash (-) in input which can replaced with underscores (_) For example, Content-Type should be replaced with content_type

A list of supported parts is available in docs for request types. examples:

  • name: Extract Server Header From HTTP Response value: > []string{"server"}
  • name: Extracting value of PHPSESSID Cookie value: > []string{"phpsessid"}
  • name: Extracting value of Content-Type Cookie value: > []string{"content_type"}

json []string

JSON allows using jq-style syntax to extract items from json response

Examples:

json:
    - .[] | .id
json:
    - .batters | .batter | .[] | .id

xpath []string

XPath allows using xpath expressions to extract items from html response

Examples:

xpath:
    - /html/body/div/p[2]/a

attribute string

Attribute is an optional attribute to extract from response XPath.

Examples:

attribute: href

part string

Part is the part of the request response to extract data from.

Each protocol exposes a lot of different parts which are well documented in docs for each request type.

Examples:

part: body
part: raw

internal bool

Internal, when set to true will allow using the value extracted in the next request for some protocols (like HTTP).


dns.Request

Request contains a DNS protocol request to be made from a template

Appears in:

extractors:
    - type: regex
      regex:
        - ec2-[-\d]+\.compute[-\d]*\.amazonaws\.com
        - ec2-[-\d]+\.[\w\d\-]+\.compute[-\d]*\.amazonaws\.com
name: '{{FQDN}}'
type: CNAME
class: inet
retries: 2
recursion: true

matchers []matchers.Matcher

Matchers contains the detection mechanism for the request to identify whether the request was successful by doing pattern matching on request/responses.

Multiple matchers can be combined with matcher-condition flag which accepts either and or or as argument.


extractors []extractors.Extractor

Extractors contains the extraction mechanism for the request to identify and extract parts of the response.


matchers-condition string

MatchersCondition is the condition between the matchers. Default is OR.

Valid values:

  • and

  • or


id string

ID is the optional id of the request


name string

Name is the Hostname to make DNS request for.

Generally, it is set to {{FQDN}} which is the domain we get from input.

Examples:

name: '{{FQDN}}'

type string

Type is the type of DNS request to make.

Valid values:

  • A

  • NS

  • DS

  • CNAME

  • SOA

  • PTR

  • MX

  • TXT

  • AAAA


class string

Class is the class of the DNS request.

Usually it's enough to just leave it as INET.

Valid values:

  • inet

  • csnet

  • chaos

  • hesiod

  • none

  • any


retries int

Retries is the number of retries for the DNS request

Examples:

# Use a retry of 3 to 5 generally
retries: 5

recursion bool

Recursion determines if resolver should recurse all records to get fresh results.


resolvers []string

Resolvers to use for the dns requests


file.Request

Request contains a File matching mechanism for local disk operations.

Appears in:

extractors:
    - type: regex
      regex:
        - amzn\.mws\.[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}
extensions:
    - all

matchers []matchers.Matcher

Matchers contains the detection mechanism for the request to identify whether the request was successful by doing pattern matching on request/responses.

Multiple matchers can be combined with matcher-condition flag which accepts either and or or as argument.


extractors []extractors.Extractor

Extractors contains the extraction mechanism for the request to identify and extract parts of the response.


matchers-condition string

MatchersCondition is the condition between the matchers. Default is OR.

Valid values:

  • and

  • or


extensions []string

Extensions is the list of extensions to perform matching on.

Examples:

extensions:
    - .txt
    - .go
    - .json

denylist []string

ExtensionDenylist is the list of file extensions to deny during matching.

By default, it contains some non-interesting extensions that are hardcoded in nuclei.

Examples:

denylist:
    - .avi
    - .mov
    - .mp3

id string

ID is the optional id of the request


max-size int

MaxSize is the maximum size of the file to run request on.

By default, nuclei will process 5 MB files and not go more than that. It can be set to much lower or higher depending on use.

Examples:

max-size: 2048

no-recursive bool

NoRecursive specifies whether to not do recursive checks if folders are provided.


network.Request

Request contains a Network protocol request to be made from a template

Appears in:

host:
    - '{{Hostname}}'
    - '{{Hostname}}:2181'
inputs:
    - data: "envi\r\nquit\r\n"
read-size: 2048
matchers:
    - type: word
      words:
        - zookeeper.version

id string

ID is the optional id of the request


host []string

Host to send network requests to.

Usually it's set to {{Hostname}}. If you want to enable TLS for TCP Connection, you can use tls://{{Hostname}}.

Examples:

host:
    - '{{Hostname}}'

attack string

Attack is the type of payload combinations to perform.

Batteringram is same payload into all of the defined payload positions at once, pitchfork combines multiple payload sets and clusterbomb generates permutations and combinations for all payloads.

Valid values:

  • batteringram

  • pitchfork

  • clusterbomb


payloads map[string]interface{}

Payloads contains any payloads for the current request.

Payloads support both key-values combinations where a list of payloads is provided, or optionally a single file can also be provided as payload which will be read on run-time.


inputs []network.Input

Inputs contains inputs for the network socket


read-size int

ReadSize is the size of response to read at the end

Default value for read-size is 1024.

Examples:

read-size: 2048

read-all bool

ReadAll determines if the data stream should be read till the end regardless of the size

Default value for read-all is false.

Examples:

read-all: false

matchers []matchers.Matcher

Matchers contains the detection mechanism for the request to identify whether the request was successful by doing pattern matching on request/responses.

Multiple matchers can be combined with matcher-condition flag which accepts either and or or as argument.


extractors []extractors.Extractor

Extractors contains the extraction mechanism for the request to identify and extract parts of the response.


matchers-condition string

MatchersCondition is the condition between the matchers. Default is OR.

Valid values:

  • and

  • or


network.Input

Appears in:


data string

Data is the data to send as the input.

It supports DSL Helper Functions as well as normal expressions.

Examples:

data: TEST
data: hex_decode('50494e47')

type string

Type is the type of input specified in data field.

Default value is text, but hex can be used for hex formatted data.

Valid values:

  • hex

  • text


read int

Read is the number of bytes to read from socket.

This can be used for protocols which expect an immediate response. You can read and write responses one after another and evetually perform matching on every data captured with name attribute.

The network docs highlight more on how to do this.

Examples:

read: 1024

name string

Name is the optional name of the data read to provide matching on.

Examples:

name: prefix

headless.Request

Request contains a Headless protocol request to be made from a template

Appears in:


id string

ID is the optional id of the request


steps []engine.Action

Steps is the list of actions to run for headless request


matchers []matchers.Matcher

Matchers contains the detection mechanism for the request to identify whether the request was successful by doing pattern matching on request/responses.

Multiple matchers can be combined with matcher-condition flag which accepts either and or or as argument.


extractors []extractors.Extractor

Extractors contains the extraction mechanism for the request to identify and extract parts of the response.


matchers-condition string

MatchersCondition is the condition between the matchers. Default is OR.

Valid values:

  • and

  • or


engine.Action

Action is an action taken by the browser to reach a navigation

Each step that the browser executes is an action. Most navigations usually start from the ActionLoadURL event, and further navigations are discovered on the found page. We also keep track and only scrape new navigation from pages we haven't crawled yet.

Appears in:


args map[string]string

Args contain arguments for the headless action. Per action arguments are described in detail here.


name string

Name is the name assigned to the headless action.

This can be used to execute code, for instance in browser DOM using script action, and get the result in a variable which can be matched upon by nuclei. An Example template here.


description string

Description is the optional description of the headless action


action string

Action is the type of the action to perform.

Valid values:

  • navigate

  • script

  • click

  • rightclick

  • text

  • screenshot

  • time

  • select

  • files

  • waitload

  • getresource

  • extract

  • setmethod

  • addheader

  • setheader

  • deleteheader

  • setbody

  • waitevent

  • keyboard

  • debug

  • sleep


workflows.WorkflowTemplate

Appears in:


template string

Template is a single template or directory to execute as part of workflow.

Examples:

# A single template
template: dns/worksites-detection.yaml
# A template directory
template: misconfigurations/aem

Tags to run templates based on.


matchers []workflows.Matcher

Matchers perform name based matching to run subtemplates for a workflow.


Subtemplates are run if the template field Template matches.


workflows.Matcher

Appears in:


name string

Name is the name of the item to match.


Subtemplates are run if the name of matcher matches.