Scan report

kserve / kserve

Vulnerability Critical

Could an attacker exploit mistakes in this code?

864 risk score

Strong warning signs. Do not run or install this without a careful review.

High
47
Medium
587
Low
301

Sources: Bandit, Semgrep, pip-audit

Malicious pattern Low

Does this code look like it is trying to cause harm?

14 risk score

Minor issues. Worth a glance, unlikely to be urgent.

High
26
Medium
3
Low
140

Sources: YARA, ClamAV, GuardDog, dependency checker

Findings overview
1,104 total findings
Likely real
98
Uncertain
21
Likely noise
985

The risk classifier rates every finding for how likely it is to be a real issue. Likely noise still counts, just much less.

What we found, in plain English

This repository has two distinct security issues that need immediate attention before it can be used in production. The most critical finding is a Cross-site scripting (XSS) vulnerability caused by untrusted user input, which means an attacker could potentially steal data or inject malicious code into web pages. Additionally, the code uses dangerous shell commands and processes files without proper validation, creating another layer of risk that attackers might exploit to run harmful scripts. While there are some signs of potential malicious intent, these appear to be standard insecure coding patterns rather than intentional attacks, so they can be addressed through routine code reviews. However, because the overall vulnerability score is extremely high at 864.0, the repository should not be deployed until these critical flaws are fixed by the development team.

Written by a local language model from the findings below. The risk scores come from the classifier, not from this summary.

Detections by engine

  • Semgrep 873

    Pattern-based code analysis · Vulnerability

  • YARA 163

    Malware signature rules · Malicious pattern

  • Bandit 62

    Python security linter · Vulnerability

  • GuardDog 6

    Supply-chain attack heuristics · Malicious pattern

Score over time

Overall score each time this repository was scanned.

Findings

Highest severity first. Select a finding to see the code and what it means.

Large result: showing the top 431 of 1,104 findings (up to 200 per engine, most severe and most likely real first). Totals above include everything.