Boilerplate

Template scanner to speed up creating new ones

The only thing scanners need to follow is:

• process a single input file that will be specified as first positional argument
• produce as many output files as wanted in /output path (inside the container)

Input file format will match whatever is chosen as Input Generator and the output produced should match whatever is expected by Output Parser <-- settings defined in the scanner configuration in Surface)

This repository is actually built as a scanner (named example) that can be used to test scanners framework or a new rootbox.

NOTE

This is a dumb scanner as it only resolves A records and it shells out to host utility. This has a performance impact (spawning shells) over simply using a DNS lib/package but the purpose is to exemplify the common scenario of building entrypoints to wrap external/3rd-party scanners

Rest of README.md is also part of the example, feel free to re-use

Example Scanner in Docker

This scanner performs DNS resolution.

Input expected: one hostname per line Output provided: single file, with hostname:IP

Usage

Mass DNS Resolver

positional arguments:
  inputfile             THE input file

optional arguments:
  -h, --help            show this help message and exit
  -o OUTPUT, --output OUTPUT
                        output directory for results
  -p PROCESSES, --processes PROCESSES
                        number of processes
  --echo                just echo input back instead of resolving DNS (passive
                        testing)
  --wait WAIT           sleep WAIT seconds after completion

echo will just echo back so it can be executed without hammering nameservers (if not required).

wait will introduce sleep at the end which can be useful to easily simulate a long scan.

Run locally

• Build: ./build.sh nopush
• Create /tmp/input/input.txt and /tmp/output.
• Run: docker run --rm -v /tmp/input:/input -v /tmp/output:/output ${PWD##*/}:latest