{"slug": "from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during", "title": "From \"Who Wrote This?\" to \"Provenance, Actioned\": Making AI-origin code obvious during review", "summary": "LineageLens has introduced actionable provenance features that surface AI-generated code context directly in code reviews, including sidebar captures, drag-and-drop insertion, and a confidence engine. The system provides reviewers with immediate access to the original prompt, model, and confidence score for any AI-produced code block, enabling triage in under five minutes instead of 30 to 120. By replacing archival audit logs with inline, one-click actions, the tool aims to reduce reverts and shorten review cycles.", "body_md": "TL;DR: The most useful provenance is actionable provenance. Instead of storing prompts like a dusty audit log, surface them where decision-makers work: the code review. Recent UX and correlation work in LineageLens — sidebar captures, drag/drop, click-to-insert, and a confidence engine — demonstrate how provenance can shorten review cycles and reduce reverts.\n\nThe problem (why it matters)\n\nBy 2026, AI is a first-class development tool. Good suggestions become accepted edits, then commits. When reviewers see unfamiliar code they ask the obvious questions: who wrote this, why was it accepted, and was it audited? Git blame shows an author, but not the conversational context that generated the code. That missing context causes three predictable costs:\n\nTime to reproduce: reviewers re-run prompts or attempt to reproduce edits.\n\nConservative reverts: unknown edits get reverted, losing useful fixes.\n\nRisk hiding: sensitive changes slip through without proper checks.\n\nWhat \"actionable provenance\" looks like\n\nActionable provenance answers reviewer questions immediately:\n\nWho/what produced this block (adapter + model)\n\nThe original prompt text\n\nConfidence that the prompt maps to the inserted code\n\nQuick actions: insert into the editor, copy prompt to PR comment, or open the capturing session\n\nMinimal latency, clear UI, and one-click actions are the difference between “archival” and “actionable”.\n\nRecent product signals that make it practical (evidence from the repo)\n\nDrag-and-drop + click-to-insert (sidebar improvements): let reviewers place the original generated block into a temporary editor buffer or paste the prompt into the PR comment box.\n\nConfidence engine and dynamic routing: correlation scores and better adapter matching reduce false positives so reviewers can rely on the provenance instead of treating it as noise.\n\nUX fixes (trash/clear buttons, reorder, inline hover actions): small changes that keep the capture panel usable during real reviews.\n\nSee the architecture summary in the repo: architecture.md:1 and the product README (README.md:1).\n\nConcrete workflow: a reviewer’s day with actionable provenance\n\nPR opens. Reviewer scans diffs.\n\nA capture badge is shown next to changed hunks indicating \"Provenance: available — confidence 88%\".\n\nClick: the capture sidebar opens to the exact prompt, model, and surrounding context snapshot.\n\nAction buttons:\n\n\"Insert at cursor\" — drop the generated block into a temp editor to run tests locally.\n\n\"Copy prompt\" — paste into a PR comment template to ask follow-ups.\n\n\"Annotate PR\" — append an auto-formatted provenance note (prompt, model, confidence).\n\nOutcome: reviewer spends <5 minutes to triage instead of 30–120.\n\nUX trade-offs and governance constraints\n\nConfidence thresholds: too low → noisy provenance, too high → missed attributions. Tune by starting conservative (show medium/high only) and lower threshold based on false-negative feedback.\n\nPrivacy and storage: prompts can be sensitive. Default to workspace-local storage and make PR annotation an explicit reviewer action.\n\nReviewer training: a short guideline (one paragraph) in your PR template — \"If provenance shows 'high confidence', prefer triage over revert\" — makes a measurable difference.\n\nQuick checklist to evaluate your team's readiness\n\nAre prompts stored in your control plane or vendor logs? If not, you may lack necessary evidence.\n\nCan you surface provenance inline in PRs or via your editor? If not, archival logs won't help reviewers.\n\nDo you have a workflow for annotating PRs with provenance? If not, create a two-line PR template snippet now.", "url": "https://wpnews.pro/news/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during", "canonical_source": "https://dev.to/pn_28428886923dfc665/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during-review-531a", "published_at": "2026-05-29 07:21:22+00:00", "updated_at": "2026-05-29 07:41:38.422284+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-tools", "ai-products", "ai-infrastructure", "mlops"], "entities": ["LineageLens"], "alternates": {"html": "https://wpnews.pro/news/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during", "markdown": "https://wpnews.pro/news/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during.md", "text": "https://wpnews.pro/news/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during.txt", "jsonld": "https://wpnews.pro/news/from-who-wrote-this-to-provenance-actioned-making-ai-origin-code-obvious-during.jsonld"}}