{ "schema_version": 1, "builder_plan": { "scenario": "An agent reads external documents or tool responses that may contain malicious instructions.", "change": "Replay attacks and legitimate lookalikes in an isolated copy of your agent workflow.", "baseline": "Your existing checks with the same tasks, permissions, and tool responses.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Successful attacks missed, legitimate actions blocked, detection delay, and review cost.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2610.10742", "title": "BRANCH: Bypassing Multi-Scanner AI Guardrails", "summary": "BRANCH replaces single-scanner adversarial perturbation methods with a branching tree search that perturbs individual scanners while optimizing against the combined guardrail verdict. It costs attack-side compute and query budget (reported as lower than prior techniques) and, for a defensive adopter, implementing equivalent evaluation requires building a multi-scanner test harness. For the reader it can fail because its results demonstrate that deployed guardrail stacks, including commercial ones, are bypassable, so any production reliance on scanner-based guardrails alone leaves residual jailbreak risk.", "rationale": "The work is an offensive technique whose practical value to a constrained-infrastructure engineer is defensive: it shows that multi-scanner guardrails cannot be treated as a sufficient control, so the reader should monitor this line of work and add defense-in-depth rather than adopt the attack method itself.", "technique": "Branching tree-search adversarial bypass of multi-scanner LLM guardrails", "family": "seguranca_e_guardrails", "recommendation": "observar", "published": "2026-10-07", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.10742", "independent_implementations": 0, "total_implementations": 1, "stars": 1, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "BRANCH replaces single-scanner adversarial perturbation methods with a branching tree search that perturbs individual scanners while optimizing against the combined guardrail verdict. It costs attack-side compute and query budget (reported as lower than prior techniques) and, for a defensive adopter, implementing equivalent evaluation requires building a multi-scanner test harness. For the reader it can fail because its results demonstrate that deployed guardrail stacks, including commercial ones, are bypassable, so any production reliance on scanner-based guardrails alone leaves residual jailbreak risk.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.10742", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "BRANCH reports 100% attack success across 6 guardrail systems in 120 scenarios with 72% fewer queries and 4.5x reduced wallclock time versus established bypass techniques, with transfers to 29 unseen guardrails including 8 commercial black-box systems.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.10742", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }