Private Beta

Your agents are spending.
Is anyone watching?

Valkurai is a financial firewall that sits between every AI agent and every payment. Identity verification, budget policy, prompt injection detection, and human approval — before a single cent moves.

Gate 1Identity
Gate 2Policy
Gate 3Rules + Intent
Gate 4Approval
Gate 5Payment
View open standard →

100 transactions free · No credit card · Beta access

5
Gates before Stripe sees anything
0
Raw keys ever stored in the database
100%
Of decisions logged immutably
88%
Of enterprises hit by AI agent incidents in 2026

Real incidents. Real money.

These are documented cases where AI agents spent money without meaningful controls. They are not hypotheticals.

$82,314

A startup's Google API key was compromised. Attackers ran Gemini API calls for 48 hours. Normal monthly spend: $180. Google cited shared responsibility and declined to cover it.

Gate 1 — stolen key, instant revocation
$47,000

Four LangChain agents entered an infinite retry loop for 11 days. The team assumed rising costs were organic growth. No spend cap. No loop detection. No circuit breaker.

Gate 2 — per-agent spending cap
$441,000

An AI agent (Lobstar Wild) suffered a session crash. On reboot, a decimal parsing error caused it to send 52 million tokens — 5% of total supply — to a random address.

Gate 4 — human approval above threshold

More documented incidents: github.com/valkurai/agent-financial-governance

Five gates. One API call.

Add Valkurai to any AI agent in minutes. Works with LangChain, CrewAI, OpenAI Assistants, Anthropic, and any HTTP-capable framework.

01 — Register

Give your agent an identity

Register your agent with a spending cap, category allowlist, and approval threshold. One API call. The key is PBKDF2-hashed and never stored in plaintext.

POST https://api.valkurai.com/v1/agents/register
02 — Screen

Replace direct payment calls

Before any payment, screen the transaction through Valkurai. Works with your existing Stripe integration — Valkurai sits in front of it, not instead of it.

POST https://api.valkurai.com/v1/request
03 — Respond

Handle three deterministic outcomes

SAFE proceeds immediately. FLAGGED routes to you for approval via Slack, email, or SMS. BLOCKED is rejected instantly. Every outcome is logged immutably.

{"status": "SAFE" | "FLAGGED" | "BLOCKED"}
04 — Audit

Prove what your agent did

Every decision — including rejections — is logged with agent identity, request payload, classification reason, and timestamp. SHA-256 integrity hash per record. 10-year retention.

GET https://api.valkurai.com/v1/audit

Three outcomes. No ambiguity.

Every transaction produces exactly one of three outcomes. No timeouts. No partial states. No silent failures.

SAFE

All gates passed

Identity verified, policy compliant, no adversarial patterns detected, below approval threshold. Stripe is called. Payment proceeds. Audit record written.

FLAGGED

Human approval required

Transaction exceeds your approval threshold. Stripe is not called. You receive a Slack, email, or SMS notification. You have 60 minutes to approve or deny. Auto-deny on timeout.

BLOCKED

Rejected automatically

Policy violation, adversarial pattern detected, or category not permitted. Stripe is not called. Reason written to audit trail. No human action required.

Built for regulated industries.

One-click export in formats accepted by auditors. Australian-hosted infrastructure for data sovereignty.

ISO 42001
AI management system audit export
EU AI Act
Art. 9, 12, 13, 14, 86 evidence trail
APRA CPS 230
Operational risk control evidence
ASD ISM
Australian government security framework
AU AI Ethics
CSIRO ethical AI framework alignment
SOC 2
Type II in progress

All data processed exclusively in AWS ap-southeast-2 (Sydney). No data leaves Australia at Phase 1.

Start free. Scale when you ship.

No credit card required for beta access. Enterprise pricing is engagement-based.

Free
$0/month
↑ 100 tx/month · always free
  • 100 transactions/month
  • 1 agent
  • Full audit log
  • Slack + email alerts
  • SMS alerts
  • Approval workflow
  • Compliance exports
Request beta access
Enterprise
$6K+/month
Engagement-based. Contact us for a scoped quote.
  • Unlimited transactions
  • Unlimited agents
  • ISO 42001 audit export
  • EU AI Act compliance report
  • RBAC + SSO (Okta, Azure AD)
  • SOC 2 documentation
  • SLA + dedicated support
Talk to us

The financial_governance spec.

Valkurai is the reference implementation of an open, vendor-neutral standard for declaring runtime financial controls in AI agent definition files.

Spec

financial_governance block

A YAML configuration block that sits under compliance in your agent definition file. Declares spending caps, approval thresholds, and which firewall enforces them. Vendor-neutral. Apache 2.0.

github.com/valkurai/gitagent-spec →
RFC

Community validated

The spec emerged from community discussion at open-gitagent/gitagent-protocol issue #38, where the absence of runtime financial enforcement was independently identified by contributors. Endorsed as technically sound.

View RFC thread →
Incidents

Living incident database

A curated, public database of real AI agent financial incidents — updated as the space evolves. Use it for CISO conversations, design partner discussions, and regulatory submissions.

View incident database →