{
  "type": "option",
  "name": "LLM-assisted review",
  "slug": "llm-assisted-review",
  "section": {
    "id": "audit",
    "number": "03",
    "title": "Auditing a ZK protocol",
    "url": "https://zkpick.com/audit/"
  },
  "url": "https://zkpick.com/audit/llm-assisted-review/",
  "markdown": "https://zkpick.com/md/audit/llm-assisted-review.md",
  "category": "Emerging tooling",
  "summary": "LLM-assisted review: Emerging tooling. Layer: Circuit / libraries. Characteristic finds: Sustained attention over large cryptographic library surfaces (evidence so far: a single self-reported finding). Blind to: Unknown and unmeasurable; false negatives are invisible. Coverage claim: None. Cost profile: Low per run; expert triage is the real cost. Choose it when: A supplementary sweep alongside manual review and deterministic tooling, especially over large cryptographic library surfaces. Not a substitute for any other category.",
  "description": "Automated review pipelines built on large language models with domain-specific context engineering, run over circuit and cryptographic library code. The category has moved from speculative to evidenced: a missing subfield membership check in an optimised pairing routine was found this way, assigned a CVE, and fixed upstream. Disclosure: that finding is the work of this document's maintainer, and it is a single self-reported result — weight it accordingly against the peer-reviewed evidence behind the other categories here.",
  "notes": [],
  "sections": [],
  "strengths": [
    "Sustains attention over large codebases where human reviewers fatigue",
    "Not restricted to a modelled bug class the way SMT and lint tooling is",
    "Cheap enough to re-run on every change"
  ],
  "tradeoffs": [
    "Results depend heavily on pipeline design; naive prompting fails where context-engineered flows succeed",
    "No soundness or coverage guarantee; false negatives are invisible",
    "False positives consume expert triage time, which is the scarce resource",
    "Leading pipelines are proprietary and not independently reproducible, so published claims cannot be checked",
    "Must not be represented to third parties as equivalent to human or formal review"
  ],
  "bestFit": "A supplementary sweep alongside manual review and deterministic tooling, especially over large cryptographic library surfaces. Not a substitute for any other category.",
  "maintainers": null,
  "maturity": "Emerging; one CVE-assigned production finding",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "source": "https://blog.zksecurity.xyz/posts/openvm-bugs/",
  "facts": [
    {
      "table": "What each assurance technique can and cannot find",
      "url": "https://zkpick.com/audit/#audit-matrix-llm-assisted-review",
      "tech": "LLM-assisted review",
      "layer": "Circuit / libraries",
      "catches": "Sustained attention over large cryptographic library surfaces (evidence so far: a single self-reported finding)",
      "blind": "Unknown and unmeasurable; false negatives are invisible",
      "coverage": "None",
      "cost": "Low per run; expert triage is the real cost"
    }
  ],
  "relatedFailureModes": [],
  "sources": [
    {
      "title": "Missing subfield membership check in OpenVM pairing — CVE-2026-46669",
      "url": "https://blog.zksecurity.xyz/posts/openvm-bugs/",
      "kind": "disclosure"
    }
  ],
  "updated": "2026-09-12",
  "version": "1.3",
  "canonical": "https://zkpick.com/audit/llm-assisted-review/",
  "authors": [
    "MarketComp"
  ]
}