{
  "added_at": "2026-10-20",
  "arxiv_id": "2609.29744",
  "arxiv_version_reviewed": "v1",
  "authors": [
    "Douglas Leith"
  ],
  "bibtex_url": "https://awesomerea.org/papers/citations/paper-extractor-development-reliability.bib",
  "code_url": null,
  "date_basis": "First arXiv submission; the reviewed version is v1, not a verified proceedings publication.",
  "date_basis_zh": "采用 arXiv 首次提交日期；审阅版本为 v1，并非已经核实的会议录出版。",
  "date_precision": "day",
  "description_en": "Studies development of the Extractor reverse-engineering tool through commits, available session logs and code-provenance tracing. Taxonomies separate instruction intent, change triggers and agent claims, exposing errors corrected before commits that ordinary repository history misses.",
  "description_zh": "通过提交、可用会话日志及代码溯源研究 Extractor 逆向工具的开发，用分类体系区分指令意图、变更触发因素及智能体陈述，揭示普通仓库历史无法反映的提交前纠错。",
  "doi": null,
  "entry_url": "https://awesomerea.org/papers/paper-extractor-development-reliability/",
  "entry_url_zh": "https://awesomerea.org/zh/papers/paper-extractor-development-reliability/",
  "evidence_urls": [
    "https://arxiv.org/abs/2609.29744",
    "https://arxiv.org/html/2609.29744v1",
    "https://github.com/doug-leith/extractor-code-development/blob/4f74f798e58073c23386a1516b582ed6827bb190/README.md",
    "https://github.com/doug-leith/ios-protobuf-decoding/blob/9a4a688c52686f82c1a86e386848edc4eb671746/README.md"
  ],
  "first_arxiv_date": "2026-09-24",
  "id": "paper-extractor-development-reliability",
  "json_url": "https://awesomerea.org/papers/paper-extractor-development-reliability/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/papers/paper-extractor-development-reliability/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/papers/paper-extractor-development-reliability/index.md",
  "metadata_notes": "The author's pinned reproducibility-archive README documents lost sessions and unreproducible response-scoring evidence. The repository exposes an archive and Git-history material. No repository-wide license was verified; the project link points to the artifact landing page.",
  "metadata_notes_zh": "作者固定版本复现归档 README 记载丢失的会话及缺少证据而无法复现的响应评分。仓库提供归档和 Git 历史材料。未核实仓库整体许可，项目链接指向产物页面。",
  "pdf_url": "https://arxiv.org/pdf/2609.29744v1",
  "project_url": "https://github.com/doug-leith/extractor-code-development",
  "publication_date": "2026-09-24",
  "publication_status": "preprint",
  "record_type": "paper",
  "runtime_tested": false,
  "scope_limitations_en": "One codebase with human-directed work, no human-coded comparison. The 14.3% figure is 15 of 105 audited generation events, including test defects and stale tests, not a defect rate per source line or in shipped code. LLM scoring has sampled human calibration; transcripts are incomplete and the archive omits per-response evidence bundles. This companion study does not independently corroborate the related schema-recovery results.",
  "scope_limitations_zh": "仅研究一个由人指导开发的代码库，没有人工编写代码的对照。14.3% 指 105 次已审计生成事件中的 15 次，包含测试缺陷及未同步更新的测试，不是源码行或已交付代码的缺陷比例。大语言模型评分接受抽样人工校准，但会话记录不完整，归档缺少逐响应评分证据包。该配套研究并不独立佐证相关模式恢复结果。",
  "targets": [
    "binary",
    "protocol"
  ],
  "title": "Between the Commits: Process, Error, and Claim Reliability in a Wholly AI-Authored Codebase",
  "title_zh": "提交之间：一个完全由 AI 编写的代码库中的过程、错误与陈述可靠性",
  "url": "https://arxiv.org/abs/2609.29744",
  "venue": "arXiv",
  "verification": "source-reviewed",
  "verified_at": "2026-10-20",
  "why_fits_en": "Adds provenance and claim-auditing methods for agent-built reverse-engineering tools.",
  "why_fits_zh": "补充智能体构建逆向工具时的溯源与陈述审计方法。",
  "year": 2026
}