Metraipolis › archive › dc_8754bbfe750a

EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents — arXiv:2608.30429

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30429
License (as archived)
CC0-1.0
Archived
2026-10-17T07:44:40.142Z
SHA-256 of the archived text
dc88326bee4ca7998b842320d92cd2442c6cea32c4f466af9a4aa6a1bbf1e626
Stable URL of this page
https://metraipolis.com/documento/dc_8754bbfe750a

How to cite

EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents — arXiv:2608.30429 — arXiv (en), https://arxiv.org/abs/2608.30429 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8754bbfe750a (Metraipolis archive dc_8754bbfe750a, sha256 dc88326bee4ca7998b842320d92cd2442c6cea32c4f466af9a4aa6a1bbf1e626)
BibTeX
@misc{dc8754bbfe750a,
  title = {EvoSkill Injection: Red-Teaming Autonomous Skill Generation and Evolution in Self-Evolving Agents — arXiv:2608.30429},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30429},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8754bbfe750a; sha256 dc88326bee4ca7998b842320d92cd2442c6cea32c4f466af9a4aa6a1bbf1e626},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30429. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM-based agent systems increasingly adopt skill-based architectures to reduce repetitive reasoning costs and improve stable, efficient task execution. Recent studies propose self-evolving agents that autonomously generate, refine, and reuse skills from past experiences to enable continuous capability evolution. However, autonomous skill evolution introduces a new attack surface in which malicious capabilities are generated, stored, and reused as legitimate skills. In this paper, we define EvoSkill Injection as a threat model targeting the autonomous skill generation and evolution pipeline of self-evolving agents. We further propose SARGE (Red-teaming Autonomous Skill Generation and Evolution in self-evolving agents), a red-teaming framework for evaluating this threat model through iterative generation, escalation, and reinforcement interactions. To support our framework, we construct EvoSkillBench, a benchmark dataset of malicious interaction trajectories for inducing malicious skill formation in self-evolving agents, and introduce EvoSkillSafetyBench, a post-attack benchmark for evaluating whether injected malicious skills are subsequently retrieved and activated as harmful behaviors. Our evaluation shows that SARGE induces malicious skill formation and that injected skills are persistently stored and repeatedly activated, highlighting the risk of persistent capability corruption.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.30429 — 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 2026-10-17T07:28:26.604Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).