Metraipolis › archive › dc_e93fcf35886b

DarKnight: Retrieval-Guided Agentic Framework for Breaking Dark Web CAPTCHA Gates — arXiv:2610.19830

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19830
License (as archived)
CC0-1.0
Archived
2026-10-15T21:46:39.598Z
SHA-256 of the archived text
ee2c358fa30bac68fdfb3c01d8830e1c1037c0665e975736c49341b7d3ad65d0
Stable URL of this page
https://metraipolis.com/documento/dc_e93fcf35886b

How to cite

DarKnight: Retrieval-Guided Agentic Framework for Breaking Dark Web CAPTCHA Gates — arXiv:2610.19830 — arXiv (en), https://arxiv.org/abs/2610.19830 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e93fcf35886b (Metraipolis archive dc_e93fcf35886b, sha256 ee2c358fa30bac68fdfb3c01d8830e1c1037c0665e975736c49341b7d3ad65d0)
BibTeX
@misc{dce93fcf35886b,
  title = {DarKnight: Retrieval-Guided Agentic Framework for Breaking Dark Web CAPTCHA Gates — arXiv:2610.19830},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19830},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e93fcf35886b; sha256 ee2c358fa30bac68fdfb3c01d8830e1c1037c0665e975736c49341b7d3ad65d0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19830. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cyber Threat Intelligence (CTI) investigators who monitor dark web marketplaces for leaked credentials and breach listings must pass the diverse CAPTCHAs that operators deploy to block automated collection. Existing solvers either need unscalable per-site fine-tuning or rely on commercial Vision-Language Model (VLM) APIs, which surrender operational sovereignty: provider policies, Tor-exit blocking, government regulation, or model updates outside the investigator's control can halt or alter the solver overnight, and per-call costs recur with every session and retry. We present DarKnight, a self-improving agentic framework designed to run on a single workstation GPU inside an investigator's Tor crawler and solve CAPTCHAs with small, locally hosted open-weight VLMs that no provider decision can withdraw, without per-market training or third-party services. For each CAPTCHA, DarKnight retrieves successful and failed past attempts on similar images (dual-signal RAG), composes an image-specific preprocessing pipeline, and learns from the site's pass/fail verdict. In a case study on four real-world dark web marketplaces, agentic tool selection outperformed parameter scaling. A deployable open-weight VLM with DarKnight reached 83.2\% on the most heavily defended market, 37.2 percentage points above its base model, outperforming even the frontier Claude Opus~5 (44%). DarKnight recovered 48.2% of baseline failures at a 7.6% degeneration cost, indicating that selective tool use can avoid much of the degradation caused by uniform preprocessing.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.19830 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T21:33:44.512Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).