Grounding & verification for AI work.
Turn your rules, facts & standards into the rails your AI runs on.
Rules, thresholds, reversals — captured as cited factlets, kept current, and applied by every agent in Claude, Cursor, and Copilot. Every output gets a verdict against them — so you can review each step yourself, or let agents run and audit the trail after.
-
01
Ground before it writes Booster supplies the facts that govern the work — the agent reasons from your truth, not the internet's average.
-
02
Verify after it writes Inspector checks every output against those same facts and returns a cited pass/fail verdict — with the rule it broke.
-
03
Open Factlet protocol Git-native, vendor-neutral, W3C PROV-O lineage. Not a proprietary format.
Engineering, Manufacturing, and Finance panels show illustrative scenarios. The ROI report is from one of our own logged sessions.
New — Kernora Agent Security. Give your AI coding agent a security baseline in 30 seconds. One MCP URL, no install.
Try it free →Ground before it writes, verify after
Stop the agent guessing about your codebase
Before your agent answers, Booster supplies the facts that govern the work — the decisions you've made, the standards you hold, and why. The model reasons from your truth, not the public internet's average. In agent sessions, facts arrive as declarative context — no menus, no prompts an autonomous agent can't answer.
Turn "looks right" into a cited verdict
After the agent produces a change, Inspector checks it against those same facts and returns pass, or fail with a machine-readable rejection code and the exact factlet it broke — reproducible, cited, auditable.
You set the dial; coverage earns the autonomy
Nora runs the ground → act → verify loop and keeps a human in it where it matters — stepping back only as the work becomes provably checkable. You set how much human stays in the loop; your verification coverage is what earns more autonomy.
Anchor every AI reply in the policy your risk committee approved last quarter.
Grounding an AI in search results stops it guessing from training data. Grounding with Factlets does the same — but the source is your organization's curated factbook of decisions, policies, and conventions, not the public web. Same proven idea, governed at the source, audit-ready for the regulator.
Before the model writes
Like a type-checker for your organization's decisions: paths that violate your factlets get pruned during reasoning, not flagged at output. Less churn, fewer dead ends, cheaper inference.
About your organization's work
Answers come from the decisions your organization actually made — not the model's guess from training data, and not a policy your company already reversed.
Every answer names its factlet
Each reply cites the factlet (f155) and its provenance — per-factlet trust and a grounding score. Every factlet also states its basis: execution-proof, human-confirmed, or agent-asserted. Ours today: 5,270 / 3 / 2 of 5,275. Most AI memory is agent-asserted and doesn't say so. Nora says so.
Your living source of truth
The factbook updates as decisions change. New ones land, reversed ones retire automatically — so the model grounds on what your organization decided this week, not what shipped last quarter.
One verification engine, three cadences.
The same factlet-grounded check, run at the rhythm the job needs — from a single pull request, to a continuous scan, to an auditor-grade export.
A verdict before a human looks
Every PR gets a pass/fail verdict. Non-conforming code carries a rejection code that maps to the rule it broke — so the gate is trusted enough to leave on.
An audit your own tooling can read
Export the whole evidence chain as a W3C PROV-O audit your security team's tooling can read. Standards-grade lineage, not a vendor PDF.
Coverage that doesn't wait
Scan the whole repository for drift against your standards, continuously — coverage that doesn't wait for the next change.
One agent learns it. The next agent starts with it.
Agent teams re-learn the same facts every session: the build flag that works, the schema rule the docs got wrong, the fix that already shipped. The factbook is the channel between your agents — not just between you and one model.
490k tokens → 15 factlets
In one of our overnight runs, an agent's ~490k-token porting session compressed into 15 factlets. The next two agents read them before reading any code.
Gotchas survive the handoff
Across a ~30-subagent, 5-iteration pipeline, later agents consumed earlier agents' hard-won mechanics verbatim — one briefing skipped ~148k tokens and 18 minutes of re-derivation.
Corrections beat documentation
A factlet overriding a stale doc rule reached every later agent the same hour it was written. The next agent verified per-file instead of trusting a script that lied.
Numbers counted from our own agents' field logs, July 2026 — the reports ship in the repo (docs/field-reports/).
No model was trained on your organization's decisions.
Every enterprise makes decisions, reverses them, supersedes them — every week. No model was trained on yours, so your AI guesses — or grounds on the version your organization already moved past. Nora captures those decisions as they evolve, grounds every AI action in the current ones, and retires the stale ones — across engineering, regulated manufacturing, and financial close alike.
The model suggests the approach your team already reversed.
A migration can't run inside a transaction — your team learned that the hard way and wrote a new rule. The model never saw the reversal, so it confidently proposes the old way. "Obvious" is exactly where decisions and training disagree.
Nora grounds every reply in the decisions your organization holds right now — and forces each answer to cite which one it used. Open standard, audit it yourself.
Last quarter's revenue-recognition rule is this quarter's SOX 404 finding.
Finance changes a rev-rec threshold; the old rule is now a material weakness. A session ends, the context window dies, and tomorrow the AI cites the superseded ASC 606 policy in a customer memo — again. Re-explaining the current rule every session isn't just wasted work; it's a controls failure waiting for the next audit.
When a decision changes, Nora retires the stale one automatically and merges the new one into the same factbook. Tomorrow's prompt starts from this week's truth — with the policy change captured in the audit log.
The store behind every answer — git-native, yours, and version-controlled with your code.
A factbook is plain markdown plus a structured index in .nora/ next to your code. Yours on day one; it syncs to your own AWS or S3. The model can't quote a decision without naming it.
From repos & sessions
Reads code, ADRs, tickets, docs, policies. Watches every Claude / Cursor / Kiro session as it happens — supervised or autonomous. Each decision gets an ID, source, decided-on date, and privacy tier.
Booster · before it writes
The current decisions, ranked by trust, go into the prompt before the AI generates. The model writes against what your team decided — Claude, GPT, Gemini, or local.
Inspector · after it writes
A conformance check reads the output back and flags where it drifts from your current decisions. Every answer cites the decision behind it — auditable, not just plausible.
Auto-consolidation
A decision enters as proposed, earns candidate after repeated use without contradiction, and you approve canonical. When a decision reverses, the stale one retires on its own.
Click any f### chip in a reply to open its decision trail.
See what it cites, why your team decided it, which decision it superseded, who applied it last. Git-tracked alongside your code.
Built on the open Factlet protocol — your factbook is portable, your tooling isn't locked to us.
Grounding changes the answer. Open methodology · N=6.
Without a factbook, three frontier models contradicted documented team decisions 12 times in 18 answers. With one: zero, across 36. Our count from the published raw run logs — recount it yourself.
The lift is narrow — and real.
It appears where your documented decision conflicts with what the model assumes. It vanishes where general knowledge already matches — the run reports those null cells too.
Try it live: the same model call, with and without an injected factlet.
Our own eval · 6 tasks · single author · 3-model judge panel · counts are ours from the raw scored runs (the repo publishes per-response rates) · no RAG or vendor-memory comparison
Methodology + raw runs on GitHub →Team-decision contradictions · 18 answers without, 36 with · lower is better
Know what your factbook doesn't cover.
A representative-query batch runs nightly. You see which questions your team will ask that your factbook can't ground — and a concrete list of decisions to add. Not a score: a punch list.
Example dashboard · illustrative figures — your real numbers replace these once Nora runs against your factbook.
of 184 representative queries hit FactSignal ≥3.
Up from 51% two weeks ago.
Bucketed histogram · 184 queries
The 0–1 bar buckets are the productive output: 70 queries your team will ask but your factbook can't ground. Click a bucket to see the queries; click a query to see the proposed-fact diff.
28 queries · click to propose facts.
Verified nightly. Falsifiable. Shareable with your team in one screenshot.
Audit-ready by construction, not by checklist.
- PROV-O lineageW3C-standard provenance on every factlet
- Per-fact egress consenta factlet doesn't leave the host without a tier you approved
- Audit log to your S3your bucket, your KMS keys, your retention policy
- On-prem deployablesidecar Docker, no Kernora cloud in the path
- SOX · ISO 27001 · GxP-friendlysupersession trail = controls evidence the auditor can read
Built for regulated manufacturing, financial close, and codebases shipping to production. Same factbook, same controls, same proof trail.
One factbook in the middle. Five sources, five surfaces, five controls.
Nora maintains the five elements of the Factlet protocol in the middle box — facts flow in from your tools, out to every model you use, with provenance and consent attached.
"Why not AGENTS.md, a vault — or the model's own memory?"
An AGENTS.md file holds rules someone must hand-maintain. Built-in AI memory accumulates inside one vendor's silo. A vault documents what your project is. Nora does a different job: it captures what you decided — and why — grounds the AI in the current decision, and checks the output against it.
Hand-written rules. Per-vendor memory.
- AGENTS.md is only as current as the last person who edited it — nobody retires the rule you reversed
- Vendor memory lives in one tool, accumulates without supersession, and can't cite where an answer came from
- Both drift silently — and neither moves with you to the next tool
Ground the current decision. Verify the output.
- Captures decisions during sessions — automatically
- Grounds the AI in the current decision, before it writes
- Verifies the output against it, after — cited and auditable
- Retires the stale ones;
nora revisefor corrections - Same factbook across Claude, GPT, and Gemini
One command. Free for one developer, grows with the company.
Free for one developer, local-only. Pro syncs to your own S3 — your bucket, your keys. Team and Enterprise add shared factbooks and written compliance commitments.
curl -fsSL https://kernora.ai/install | bash
Registers Nora as a local MCP server for Claude Code & Claude Desktop. Restart your agent and the factbook tools appear. Verified local-only: 0 packets to kernora.ai under tcpdump.
MCP server · plugin coming
The command above registers Nora today. A Claude Code plugin is coming to the marketplace — one-step install when it lands.
Pro adds an opt-in hosted endpoint at mcp.kernora.ai for browser-based agents — off by default.
VS Code · Cursor · Kiro
One-click for VS Code from the marketplace; Cursor and Kiro side-load the same extension. A starter factbook is seeded from your repo. View the listing →
Autonomous agents in production — early access
Coming — early access. Sidecar to your agent runtime — CI bots, scheduled agents, customer-facing agents. Headless MCP, BYOK secret management, audit log to your S3, decisions auto-promoted by trust ladder. Per-fact egress consent, encrypted at rest, your bucket and keys throughout.
Compliance-ready, starts free for one developer. Enterprise tier adds compliance commitments in writing, shared governance, and dedicated support. Team shares a factbook across the org. Pro syncs to your own S3. Free covers a single developer, local only. See enterprise pricing →
Why we built this.
I was shipping a product with Claude Code, Kiro, and Cursor — sometimes three agents in parallel. Hundreds of sessions, thousands of decisions. And the same thing kept happening: I'd make a call on Tuesday — drop a library, pick a pattern, reverse an approach — and on Friday the AI would confidently suggest the exact thing I'd already rejected. The models got better every quarter — but a sharper model still has no idea what your team decided last week. So every session, the AI worked from the version we'd already moved past.
This isn't a model-capability problem. The METR study found experienced developers are measurably slower with AI — the cost of re-establishing what the organization already decided eats the gain. What holds real enterprises back isn't a weaker model — it's that the decisions they've made live outside it.
Every decision your organization makes should ground the next AI action — automatically. So we built the layer that does it.
Built products and led teams at Amazon. Built startups. Now building the grounding, verification & orchestration layer for AI — so Fortune 500 organizations can govern AI without slowing it down. See, govern, and trust how AI works inside the business.