njsscan is a static application testing (SAST) tool that can find insecure code patterns in your node.js applications using a simple pattern matcher from libsast and syntax-aware semantic code pattern search tool semgrep.
Changelog v0.2.3
Removing a leading statement ellipsis trims time on a test corpus from
32.7 to 24.2 seconds.
Bump libsast, skip files > 25MB for pattern matcher and choice matcher.
Installation
pip3 install njsscan
Use
$ njsscan
usage: njsscan [-h] [--json] [-o OUTPUT] [--missing-controls] [-v]
[path [path ...]]
positional arguments:
path Path can be file(s) or directories with source code
optional arguments:
-h, --help show this help message and exit
--json set output format as JSON
-o OUTPUT, --output OUTPUT
output filename to save the result
--missing-controls enable missing security controls check
-v, --version show njsscan version
Example
$ njsscan xss_node.js
- Pattern Match ████████████████████████████████████████████████████████████ 1
- Semantic Grep ████████████████████████████████████████████████████████████ 53
======================================================================================================
RULE ID: express_xss
OWASP: A1: Injection
CWE: CWE-79: Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting')
DESCRIPTION: Untrusted User Input in Response will result in Reflected Cross Site Scripting Vulnerability.
SEVERITY: ERROR
======================================================================================================
__________________FILES___________________________
File: xss_node.js
Match Position: 5 - 37
Line Number(s): 5: 6
Match String: var html = "Hello" + req.query.name + ". How are you?"
res.write('Response</br>' + html);
Python API
>>> from njsscan.njsscan import NJSScan
>>> node_source = '/node_source/true_positives/sqli_node.js'
>>> scanner = NJSScan([node_source], json=True, check_controls=False)
>>> scanner.scan()
{
'templates': {},
'nodejs': {
'node_sqli_injection': {
'files': [{
'file_path': '/node_source/true_positives/sqli_node.js',
'match_position': (1, 24),
'match_lines': (4, 11),
'match_string': 'var employeeId = req.foo;\n\nvar sql = "SELECT * FROM trn_employee WHERE employee_id = " + employeeId;\n\n\n\nconnection.query(sql, function (error, results, fields) {\n\n if (error) {\n\n throw error;\n\n }\n\n console.log(results);'
}],
'metadata': {
'owasp': 'A1: Injection',
'cwe': "CWE-89: Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection')",
'description': 'Untrusted input concatinated with raw SQL query can result in SQL Injection.',
'severity': 'ERROR'
}
}
},
'errors': []
}
Github Action
Add the following file at .github/workflows/njsscan.yml:
Suppress Findings
You can suppress findings from javascript source files by adding the comment //ignore: rule_id1, rule_id2 to the line that triggers the findings.
Example:
app.get('/some/redirect', function (req, res) {
var target = req.param("target");
res.redirect(target); //ignore: express_open_redirect
});
