{"slug": "from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron", "title": "From AI logger to AI governance layer, what 5,600 downloads taught us about Chron", "summary": "Chron, an MCP server that logs AI coding sessions to a tamper-evident SQLite database, has reached 5,600 installs. The project is evolving from a logging tool to an AI governance layer, adding an Attention Score to help developers triage sessions by secret exposure, sensitive code changes, and compliance findings.", "body_md": "[When we shipped Chron six months ago, the pitch was simple: ](https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fkoealdbv4xzrx50ejkgp.png)*every AI conversation should leave a timestamped, tamper-evident record.* Developers installing AI coding tools had no answer to the question \"what did the AI actually do?\" Chron was that answer.\n\n5,600 installs later, the question has changed.\n\nIt's no longer \"did we log it?\" — most teams using Chron have solved that. The new question is: **which sessions should I look at first?**\n\nThat question is harder, and it's where Chron is heading next.\n\nChron is an MCP server that runs locally alongside Claude Code, Cursor, Windsurf, and Codex. Every message, every file change, every tool call gets logged to a local SQLite database with a precise timestamp and a tamper-evident hash chain.\n\n```\nchron history\n8f3a7c9e  claude-code  API refactor         2026-07-19  14 events\nb91c2a11  cursor        Contract review       2026-07-18  8 events\n3d0a5f22  claude-code  Auth middleware fix   2026-07-17  31 events\nchron verify 8f3a\n# ✓ hash chain intact — 14 events, no tampering detected\n```\n\nOn top of that, `chron review`\n\nruns a deterministic compliance analysis across four frameworks:\n\n```\nchron review --framework=soc2 --output=evidence.html\n```\n\nThe output is a printable HTML evidence package — not a compliance opinion, but session evidence a control owner can actually use.\n\nOne of the things we deliberately got right: Chron doesn't claim to \"cover\" a framework. It covers specific controls that are observable from AI session activity. Everything else — policy documentation, physical access, contractual obligations — requires evidence from other sources.\n\n```\nchron review --framework=nist-ai-rmf --full-map\nNIST AI RMF — Full Control Coverage Map\n\n  Covered by session evidence (12)\n  ────────────────────────────────\n  GOVERN 1.1   AI policy and governance changes    rule: nist-ai-rmf.govern.ai_policy_change\n  MAP 3.5      Sensitive data in AI sessions       rule: nist-ai-rmf.map.sensitive_data_in_session\n  MEASURE 2.7  AI monitoring modifications         rule: nist-ai-rmf.measure.ai_monitoring_change\n  ...\n\n  Needs additional evidence (28)\n  ────────────────────────────────\n  GOVERN 2.1   Organizational roles defined        source: policy_docs\n  MAP 2.2      Scientific findings documented      source: policy_docs\n  ...\n\n  Manual review required (18)\n  ────────────────────────────────\n  GOVERN 3.1   Feedback mechanisms in place        human judgement required\n  ...\n\n  Out of scope (8)\n  ────────────────────────────────\n  GOVERN 5.2   Organizational policies current     not observable from session logs\n```\n\n66 NIST AI RMF subcategories. 12 covered by Chron session evidence. 54 classified honestly. The disclaimer at the top of every report: **\"Covered does not mean compliant.\"**\n\nAt 5,600 installs, the missing piece is triage.\n\nWhen you have 200 sessions logged, you don't need a list — you need a signal. Which of these sessions involved something worth a second look?\n\nThe next release adds an **Attention Score** to every session: a deterministic 0–100 score computed from three signal categories.\n\n**Secret exposure** — was a credential, API key, or token detected in the session? (+30)\n\n**Sensitive code changes** — did the AI modify auth, production config, infrastructure, or AI oversight code? (up to +30, based on path patterns)\n\n**Open compliance findings** — does this session have unresolved findings from `chron review`\n\n? Severity-weighted, capped at +30.\n\nThe result is visible everywhere you already look:\n\n```\nchron history\n8f3a7c9e  claude-code  API refactor         [ATTENTION: 86]  ●●●●\n3d0a5f22  claude-code  Auth middleware fix  [ATTENTION: 72]  ●●●○\nb91c2a11  cursor        Contract review       [ATTENTION: 12]  ●○○○\nchron risk --since=30d\nSessions requiring attention  (last 30 days)\n\n  1.  8f3a7c9e  claude-code  API refactor         Score: 86\n      ↳ secret detected in session\n      ↳ auth code modified (3 files)\n      ↳ 2 unresolved SOC 2 findings (1 critical)\n\n  2.  3d0a5f22  claude-code  Auth middleware fix  Score: 72\n      ↳ access control code modified\n      ↳ production config changed\n      ↳ 1 unresolved ISO 27001 finding\n```\n\nNo AI in the scoring path. Pure path matching + database queries. Same session, same score, every time.\n\nThe compliance question used to be \"do we have logs?\" Most teams can answer that today.\n\nThe question in 2026 is \"did the AI touch anything that warrants a review?\" That's a triage problem, not a logging problem. Attention Score is the answer.\n\nAfter that: a static local dashboard — `chron dashboard`\n\n— that generates an HTML snapshot with session list, scores, top reasons, open findings, and framework coverage. No server, no cloud account, no new dependencies. Same personality as the rest of Chron.\n\n**Chron is available on npm:**\n\n```\nnpm install -g chron-mcp\nchron-mcp   # auto-configures Claude Code, Cursor, Windsurf\n```\n\nSource and docs: [github.com/SirinivasK/chron](https://github.com/SirinivasK/chron)\n\n*Chron logs every AI conversation to a local SQLite database you own. Nothing leaves your machine.*", "url": "https://wpnews.pro/news/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron", "canonical_source": "https://dev.to/sirinivask/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron-1iek", "published_at": "2026-07-27 18:06:28+00:00", "updated_at": "2026-07-27 18:31:59.891673+00:00", "lang": "en", "topics": ["ai-safety", "developer-tools", "ai-agents"], "entities": ["Chron", "Claude Code", "Cursor", "Windsurf", "Codex"], "alternates": {"html": "https://wpnews.pro/news/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron", "markdown": "https://wpnews.pro/news/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron.md", "text": "https://wpnews.pro/news/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron.txt", "jsonld": "https://wpnews.pro/news/from-ai-logger-to-ai-governance-layer-what-5600-downloads-taught-us-about-chron.jsonld"}}