meg is a tool for fetching lots of URLs but still being 'nice' to servers.
It can be used to fetch many paths for many hosts; fetching one path for all hosts before moving on to the next path and repeating.
You get lots of results quickly, but non of the individual hosts get flooded with traffic.
meg is written in Go and has no run-time dependencies. If you have Go 1.9
or later installed and configured you can install meg with go get
:
▶ go get -u github.com/tomnomnom/meg
Or download a binary and
put it somewhere in your $PATH
(e.g. in /usr/bin/
).
If you see an error like this it means your version of Go is too old:
# github.com/tomnomnom/rawhttp
/root/go/src/github.com/tomnomnom/rawhttp/request.go:102: u.Hostname undefined (
type *url.URL has no field or method Hostname)
/root/go/src/github.com/tomnomnom/rawhttp/request.go:103: u.Port undefined (type
*url.URL has no field or method Port)
/root/go/src/github.com/tomnomnom/rawhttp/request.go:259: undefined: x509.System
CertPool
You should either update your version of Go, or use a binary release for your platform.
Given a file full of paths:
/robots.txt
/.well-known/security.txt
/package.json
And a file full of hosts (with a protocol):
http://example.com
https://example.com
http://example.net
meg
will request each path for every host:
▶ meg --verbose paths hosts
out/example.com/45ed6f717d44385c5e9c539b0ad8dc71771780e0 http://example.com/robots.txt (404 Not Found)
out/example.com/61ac5fbb9d3dd054006ae82630b045ba730d8618 https://example.com/robots.txt (404 Not Found)
out/example.net/1432c16b671043271eab84111242b1fe2a28eb98 http://example.net/robots.txt (404 Not Found)
out/example.net/61deaa4fa10a6f601adb74519a900f1f0eca38b7 http://example.net/.well-known/security.txt (404 Not Found)
out/example.com/20bc94a296f17ce7a4e2daa2946d0dc12128b3f1 http://example.com/.well-known/security.txt (404 Not Found)
...
And save the output in a directory called ./out
:
▶ head -n 20 ./out/example.com/45ed6f717d44385c5e9c539b0ad8dc71771780e0
http://example.com/robots.txt
> GET /robots.txt HTTP/1.1
> Host: example.com
< HTTP/1.1 404 Not Found
< Expires: Sat, 06 Jan 2018 01:05:38 GMT
< Server: ECS (lga/13A2)
< Accept-Ranges: bytes
< Cache-Control: max-age=604800
< Content-Type: text/*
< Content-Length: 1270
< Date: Sat, 30 Dec 2017 01:05:38 GMT
< Last-Modified: Sun, 24 Dec 2017 06:53:36 GMT
< X-Cache: 404-HIT
<!doctype html>
<html>
<head>
Without any arguments, meg will read paths from a file called ./paths
,
and hosts from a file called ./hosts
. There will also be no output:
▶ meg
▶
But it will save an index file to ./out/index
:
▶ head -n 2 ./out/index
out/example.com/538565d7ab544bc3bec5b2f0296783aaec25e756 http://example.com/package.json (404 Not Found)
out/example.com/20bc94a296f17ce7a4e2daa2946d0dc12128b3f1 http://example.com/.well-known/security.txt (404 Not Found)
You can use the index file to find where the response is stored, but it's
often easier to find what you're looking for with grep
:
▶ grep -Hnri '< Server:' out/
out/example.com/61ac5fbb9d3dd054006ae82630b045ba730d8618:14:< Server: ECS (lga/13A2)
out/example.com/bd8d9f4c470ffa0e6ec8cfa8ba1c51d62289b6dd:16:< Server: ECS (lga/13A3)
If you want to request just one path, you can specify it directly as an argument:
▶ meg /admin.php
meg's help output tries to actually be helpful:
▶ meg --help
Request many paths for many hosts
Usage:
meg [options] [path|pathsFile] [hostsFile] [outputDir]
Options:
-c, --concurrency <val> Set the concurrency level (defaut: 20)
-d, --delay <val> Milliseconds between requests to the same host (defaut: 5000)
-H, --header <header> Send a custom HTTP header
-r, --rawhttp Use the rawhttp library for requests (experimental)
-s, --savestatus <status> Save only responses with specific status code
-v, --verbose Verbose mode
-X, --method <method> HTTP method (default: GET)
-g, --saveGrep <string> Save only if grep string in response
Defaults:
pathsFile: ./paths
hostsFile: ./hosts
outputDir: ./out
Paths file format:
/robots.txt
/package.json
/security.txt
Hosts file format:
http://example.com
https://example.edu
https://example.net
Examples:
meg /robots.txt
meg -s 200 -X HEAD
meg -c 30 /
meg hosts.txt paths.txt output
By default meg will attempt to make 20 concurrent requests. You can change that
with the -c
or --concurrency
option:
▶ meg --concurrency 5
It's not very friendly to keep the concurrency level higher than the number of hosts - you may end up sending lots of requests to one host at once.
By default meg will wait 5000 milliseconds between requests to the same host.
You can override that with the -d
or --delay
option:
▶ meg --delay 10000
Warning: before reducing the delay, ensure that you have permission to make large volumes of requests to the hosts you're targeting.
You can set additional headers on the requests with the -H
or --header
option:
▶ meg --header "Origin: https://evil.com"
▶ grep -h '^>' out/example.com/*
> GET /.well-known/security.txt HTTP/1.1
> Origin: https://evil.com
> Host: example.com
...
If you want to send requests that aren't valid - for example with invalid URL encoding - the Go HTTP client will fail:
▶ meg /%%0a0afoo:bar
request failed: parse https://example.org/%%0a0afoo:bar: invalid URL escape "%%0"
You can use the -r
or --rawhttp
flag to enable use of the rawhttp
library, which does little to no validation on the request:
▶ meg --verbose --rawhttp /%%0a0afoo:bar
out/example.com/eac3a4978bfb95992e270c311582e6da4568d83d https://example.com/%%0a0afoo:bar (HTTP/1.1 404 Not Found)
The rawhttp
library and its use is experimental. Amongst other things it doesn't
yet support chunked transfer encoding, so you may notice chunk lengths interspersed
with your output if you use it.
If you only want to save results that returned a certain status code, you can
use the -s
or --savestatus
option:
▶ meg --savestatus 200 /robots.txt
You can specify which HTTP method to use with the -X
or --method
option:
▶ meg --method TRACE
▶ grep -nri 'TRACE' out/
out/example.com/61ac5fbb9d3dd054006ae82630b045ba730d8618:3:> TRACE /robots.txt HTTP/1.1
out/example.com/bd8d9f4c470ffa0e6ec8cfa8ba1c51d62289b6dd:3:> TRACE /.well-known/security.txt HTTP/1.1
...