github.com / protectai

protectai/vulnhuntr

Python·15 files·commit ead88c5·scanned 4m ago·cached ✓

25/100

LOW RISK

Minor findings consistent with the project type. Nothing reachable from install hooks.

score capped at 25 — 2,775 stars — findings likely legitimate code patterns

verdict accurate?

Research / educational context

threat-state: lowlive

FINDINGS ░▒▓

criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.vulnhuntr/prompts.py

1. High-Risk Functions and Methods:
   - eval(), exec(), subprocess modules
   - os.system(), os.popen()
   - pickle.loads(), yaml.load(), json.loads() with custom decoders

criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.vulnhuntr/prompts.py

1. High-Risk Functions and Methods:
   - eval(), exec(), subprocess modules
   - os.system(), os.popen()
   - pickle.loads(), yaml.load(), json.loads() with custom decoders

warningPython subprocess / os.system usageos.system() or subprocess with shell=True / suspicious commands can execute arbitrary shell commands. Plain pip/package installs are usually benign.vulnhuntr/prompts.py

   - eval(), exec(), subprocess modules
   - os.system(), os.popen()
   - pickle.loads(), yaml.load(), json.loads() with custom decoders

infoSuspicious files are not reachable from entry pointsFlagged files exist but are not imported by any entry point. They may be dead code, tests, or attack payloads triggered by another mechanism.+2

warningFlattened or dead control flowDetected switch(true), dead if branches, or deeply nested ternaries — patterns used by obfuscators to hide execution order.vulnhuntr/__main__.py+3

Scores are heuristics. A “safe” verdict means no known-malicious patterns were found — clever malware can look boring. Wrong verdict? Flag it above; confirmed false positives become regression tests.

5 files scanned @ ead88c5 | 9/25/2026 | heuristic scan — always review manually

risk by category

code execution21

network & exfiltration0

file system access0

obfuscation3

supply chain0

owasp / injection0

telemetry

files 5/15rules hit 5engine v6commit ead88c5

github

protectai/vulnhuntr

Zero shot vulnerability discovery using LLMs

Python

2775

320

709d

15 files

5 scanned(33%)

ead88c5

architecture░▒▓

entry (0) flagged (2) pkg (21)

25 nodes · 25 edgesscroll to zoom · click node to jump to finding