Skip to content
New

DeepSeek V4.1 Flash free for 7 days: 10M tokens a day after a top-up of $10 or more. Until Nov 30.

How it works
aeondave/adversarial-robustness-toolboxAI, agents and MCP

adversarial-robustness-toolbox

⁨Build or review adversarial-ML experiments with Adversarial Robustness Toolbox (ART): estimator adapters, evasion, poisoning, membership inference, or model extraction.⁩

v2026.11.13-21cd5683 installs+3 in 7 days39 StarsMITsynced 11 hours ago

Install

No NotHoly CLI yet? Install it first.

$ curl -fsSL https://notholy.com/install.sh | sh
$ notholy skills install aeondave/adversarial-robustness-toolbox

Writes to ⁦~/.notholy/skills⁩ after checking the signature and the SHA-256.

SignatureSigned, ed25519
cptC+wyofl…4SQzqrDw==The CLI checks the SHA-256 and the signature before it writes a file.
Packageba0a8279…323efe80
ba0a82791b9e…71a8323efe805 files · ⁦10.1 KB⁩
ScanMedium4 findings
  • Executable script shipped with the skillassets/cpu_evasion_smoke.py
  • injection.no_confirmationreferences/estimator-contract.md:3
  • injection.no_confirmationreferences/evasion-evaluation.md:40
See the full scan
Sourcegithub.com/AeonDave/malskill/tree/main/ai/adversarial-robustness-toolbox@21cd568

Use ART to implement a bounded experiment against an explicitly authorized model and dataset; ART is an API, not evidence that a model is robust or vulnerable. Pin the installed ART version and check the matching source before relying on API behavior. This guide is grounded in ART 1.20.1, tag commit cbc47a890a191809baa292992761030e7f46ab18.

Workflow and evidence contract

  1. Define model access, attacker knowledge, eligible inputs, data splits, perturbation or query budget, and success condition before choosing an attack.
  2. Wrap the actual model with an ART estimator that exposes the interfaces required by the selected attack. Verify input domain, shape, labels, outputs, loss/gradient behavior, and clipping in that wrapper.
  3. Keep training, tuning, and final evaluation data distinct. Record ART and dependency versions, estimator/backend, attack parameters, seeds, and any query accounting.
  4. Check generated data against finite-value, shape, domain/clip, and stated perturbation-budget invariants. Report clean performance and attack metrics with explicit numerators and denominators. A result from one model, split, seed, or attack is not a general robustness claim.

For each report, include: (a) art.__version__ and relevant framework versions; (b) model/estimator type and input/output contract; (c) preprocessing and clip domain; (d) threat model, attack/access mode, norm and epsilon units or query/training-data budget; (e) clean baseline, metric definition and denominator, attack parameters, seed, measured perturbation/query counts, and limitations. Separate source-established API facts from behavior observed in the run.

Load the focused reference

  • Adapter selection, preprocessing, loss gradients, output scores/logits, clipping, or an ART version change: references/estimator-contract.md.
  • White-box evasion with FGM/PGD or a defensible perturbation/query-budget metric: references/evasion-evaluation.md.
  • Authorized poisoning, membership-inference/privacy, or model-extraction workflow: references/privacy-poisoning-extraction.md.
  • Reproducible CPU FGM/PGD API and invariant smoke test, without downloads or external models: assets/cpu_evasion_smoke.py.

For challenge artifacts and challenge-specific success oracles, load AI/ML CTF instead of copying its broader playbooks here; load this skill only for ART-specific adapter or API work.

Files in this version

Permissions

PermissionDeclaredFound by the scan
Reads filesNoNo
Writes filesYesYes
Runs shell commandsYesYes
Uses the networkYesYes
Uses connectorsNoNo
Drives a browserNoNo

Scan findings

Reviewed on Nov 13, 2026

  1. MediumFile · Executable script shipped with the skillassets/cpu_evasion_smoke.py
  2. MediumPrompt injection · injection.no_confirmationreferences/estimator-contract.md:3
  3. MediumPrompt injection · injection.no_confirmationreferences/evasion-evaluation.md:40
  4. InfoLink · Links to outside domains

A clean scan lowers the risk but does not prove a package is safe. Read what you install.

VersionPublishedCommitSHA-256SignaturePermission change
v2026.11.13-21cd568Nov 13, 202621cd568ba0a8279Signed–

Ask anything your scope allows.

Create an account, add credits and ask the first question in the chat or the terminal.

  1. 1/3

    Create the account

    Individual or company. You can switch to a company later without losing balance or keys.

    IndividualCompany
  2. 2/3

    Add credits

    By card. Pay per token, no subscription.

    $5.00minimum deposit
  3. 3/3

    Ask

    In the chat, or install the CLI and run notholy login in your terminal.

    $ notholy login