# protectai/ai-exploits

A collection of real world AI/ML exploits for responsibly disclosed vulnerabilities

Repository: https://github.com/protectai/ai-exploits
Canonical: https://ross.abutalabs.com/products/ai-exploits
Language: Python
License: NOASSERTION
License Family: other
Last push: 2024-10-23T20:40:54+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 74
- inputs: {"age_days": 1043, "days_push": 679, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1746, forks 165 (observed 2026-08-28T04:05:30.772379+00:00)

## What it is
A collection of real-world AI/ML exploits and scanning templates for responsibly disclosed vulnerabilities in machine learning tools and infrastructure. It includes Metasploit modules, Nuclei templates, and CSRF templates targeting vulnerable ML components like Ray.

## Use cases
- scan servers for known AI/ML infrastructure vulnerabilities
- test ML deployment security with metasploit modules
- find vulnerable machine learning tools in my environment
- learn what real attacks against ML pipelines look like
- check if my Ray cluster is exploitable
- build a security testing toolkit for AI systems

## When to choose
- you are a security professional assessing AI/ML infrastructure
- you want Nuclei templates to scan for known ML tool CVEs
- you need to demonstrate practical AI supply-chain and infrastructure attacks to raise awareness

## When to avoid
- you need defensive patches or fixes rather than exploit code
- you are looking for a general-purpose vulnerability scanner unrelated to ML
- you expect production-ready tooling with support guarantees

## Facets
- artifact type: dataset
- maturity: active
- function: penetration-testing, vulnerability-scanning, security
- domain: security, machine-learning, penetration-testing, artificial-intelligence
- platform: python, cli
- tags: ai-security, exploits, metasploit-modules, nuclei-templates, ml-infrastructure, vulnerability-disclosure, docker

## Member repositories
- protectai/ai-exploits (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:30.772379+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:29:12.480486+00:00, confidence not recorded.
  - readme: https://github.com/protectai/ai-exploits (fetched 2026-08-28T04:05:30.772379+00:00, sha f4dac1ec2121)
- Data as of 2026-08-30T08:39:29.467469+00:00.