Verity by ProvenanceCode

Git-layer governance for AI-generated code.

Every AI coding agent — Cursor, Copilot, Devin — is pushing code into your repositories right now. Verity intercepts every AI-generated PR at the Git layer, creates a signed provenance record, and enforces your policy before anything merges.

GitHub Marketplace · planned
ISO 42001 policy
Zero-config Git integration
Verity · PR policy engine
git push origin feature/ai-refactor
Verity: intercepting PR #418...
Creating DEO · AI source: Cursor
Files changed: 12 · Risk: MEDIUM
Policy: auth_scope_changed = GATE
DEO signed. Review required.
Provenance record committed to git history.
// PR blocked pending lead engineer approval

Consequential AI-generated changes become a Developer Execution Object.

Routine commits flow through automatically. When an AI-generated change touches auth logic, payment paths, data access, or your defined risk thresholds — Verity creates a DEO: a signed, immutable record capturing what changed, which model generated it, the policy verdict, and who approved or blocked it.

📄
AI source attribution

Every DEO records which AI tool (Cursor, Copilot, Devin) generated the code and which model version was active at the time.

⚖️
Policy verdict

Policy rules are deterministic — not another LLM call. Low-risk changes auto-pass with zero friction. Only changes that cross your defined thresholds require human sign-off.

🔒
Immutable git record

Signed provenance stored directly in your git history. Auditable by compliance, legal, or security teams without leaving the repository.

👥
Human-in-the-loop

Gated PRs notify the designated reviewer. The AI-generated change cannot merge until a human engineer explicitly approves the DEO.

Three verdicts. One standard.

AUTO ✓
Auto-approve

Low-risk changes (docs, tests, cosmetic) pass automatically. DEO created, signed, and committed. Zero developer friction.

GATE ⚠
Require review

Medium-risk changes (auth, data access, API changes) trigger a required review. PR is blocked until a designated engineer approves the DEO.

BLOCK ✗
Block merge

High-risk changes (credential exposure, policy violations, out-of-scope modifications) are blocked entirely. Cannot merge without escalated override.

Built for engineering teams shipping with AI.

CTO · Head of Engineering

You've adopted AI coding tools across the team. You need governance before a security incident forces the conversation. Verity gives you policy enforcement without slowing the team down.

See CTO view →
Lead Engineer · Staff Engineer

You're the one reviewing AI-generated PRs manually today. Verity automates the triage, flags the risky ones, and gives you a signed audit trail — so your reviews are faster and defensible.

See Engineer view →

What you can buy.

ProvenanceCode is modular. Start with what you need today and expand as your AI deployment grows. No pricing listed yet — we size engagements on scope. Talk to us first.

Start here
Verity

Git-layer governance for AI-generated code. GitHub App, DEO pipeline, PR gating, policy-as-code, compliance dashboard. Private preview — request access.

⚙️
Agent OS Core

Runtime governance for AI agent actions. Policy Gateway, Executor layer, provenance records, AWS Marketplace deployment. For teams running autonomous agents in production.

🏢
Enterprise deployment

Air-gapped or private-cloud deployment of the full ProvenanceCode stack. Custom policy authoring, dedicated onboarding, SLA, and audit reporting for regulated industries.

Request early access →

Request access. Know exactly what happens next.

01
You submit the request form
Takes two minutes. Tell us your repo setup, how many engineers are using AI coding tools, and whether you have existing policy requirements (SOC 2, ISO 27001, EU AI Act).
02
We respond within one business day
A founder or engineer — not a BDR — reviews your request and replies. We'll confirm fit or suggest a call if the setup is more complex.
03
You get the GitHub App install link
Install on your repositories, connect your policy file (or use the default ISO 42001 preset), and Verity is intercepting PRs within minutes. No infrastructure.
04
First DEO generated on your first AI PR
You'll see the verdict (AUTO / GATE / BLOCK), the signed provenance record, and the audit trail — all on the first PR that hits the gate. Test in a non-production repo first if you prefer.
Request early access →
No pricing conversation required. No sales deck. Direct response from the team.

Same standard. Two enforcement surfaces.

Verity governs AI-generated code at the Git layer. Agent OS governs AI agent actions at runtime. Both products share the same ISO 42001 policy engine and provenance standard — so your audit trail is consistent across the full AI stack.

Get early access to Verity Learn about Agent OS →