🔒 All inputs are processed locally in your browser. No data is transmitted. Do not enter real personal data — use synthetic or anonymised inputs only.
OpenChainGraph · ART-15 · Agentic · AP2 · Mandate Sandbox

Agentic Mandate Sandbox

Simulate and export agent payment policies for tokenized A2A corridors. Configure spend caps, MCC allowlists, velocity throttles, and approval thresholds — test against synthetic transactions before real money moves. OpenChainGraph node ART-15 — emits a §4 hash-anchored artifact for chain composition.

v1.0.0 · §4 hash-anchored
✓ Zero Network Transit ✓ Zero PII AP2-Compatible Export Synthetic Simulation §4 Hash-Anchored OpenChainGraph ART-15
Natural-Language Policy Draft
Spend Caps
$5,000
Scope & reliance — All inputs processed locally. No data transmitted. Use synthetic inputs only. Embedded thresholds are static reference values — verify against current primary sources before acting. Deterministic · no inference · zero PII · CC BY 4.0 · OpenChainGraph ART-15.
$15,000
$100,000
$2,000
MCC Allowlist

Click to toggle allowed merchant category codes

Velocity Limits
txns/hr
txns/day
min between txns
Allowed Time Windows
Allow Weekend Transactions
Sat–Sun UTC
Block Holiday Calendar
Jurisdiction-specific
Approval & Multi-Sig
Single-Sig Auto-Approve
Below flag threshold
Dual-Sig Required
Above flag threshold
Board Approval Gate
Above daily cap × 50%
Rail & Corridor
Cross-Border KYC Check
Validate corridor compliance
OFAC / Sanctions Screen
Counterparty screening
Simulation Engine
20 txns
⚡

Configure your mandate policy on the left, then click Run Simulation to generate synthetic transactions and evaluate them against your rules.

🧠

Run a simulation to see plain-English explanations of why each transaction was approved, flagged, or denied.

// Mandate JSON will appear here after running a simulation or clicking "Generate Mandate"
// Configure the policy panel and run the simulation, or import an existing mandate.

The Agentic Mandate Sandbox lets treasury teams define agent payment policies — spend caps, MCC allowlists, velocity throttles, approval thresholds, and time windows — then test them against synthetic transactions before real money moves. Outputs are directly compatible with the AP2 mandate schema. OpenChainGraph node ART-15 — emits a §4 hash-anchored artifact for chain composition in the agentic-policy chain.

How it works: Configure the policy panel manually or paste a natural-language policy description that the tool parses into structured config. The synthetic transaction generator creates realistic test payloads across five distribution modes. A rule-based evaluator returns Approved / Flagged / Denied decisions with plain-English reasoning for each transaction. Export the resulting §4 OpenChainGraph artifact for ingestion by MCP or agent runtimes.

All logic runs as plain JavaScript in your browser — no machine learning models, no external API calls. Outputs are generated deterministically from embedded rule sets.

Export: Results can be downloaded as JSON or Markdown for use in downstream workflows, agent runtimes, or audit documentation.

Limitations: This tool is a reference aid and simulation environment. It does not constitute legal, financial, or compliance advice. Always verify outputs with qualified professionals before acting on them.

This tool runs entirely in your browser. No data is transmitted to any server. All inputs, calculations, and generated outputs are processed locally on your device. fetch() and XMLHttpRequest are overridden at runtime to block all outbound network calls. Any configuration you choose to save is stored exclusively in-memory (cleared when you close the tab) and never leaves your device.

No analytics, no cookies, no telemetry, no tracking pixels. This tool does not collect, store, or transmit any personally identifiable information (PII), financial data, or organisational data.