xBPP (Execution Boundary Permission Protocol) is the open standard for governing autonomous AI agent payments. It provides a chain-agnostic governance layer that evaluates agent spending intent against declarative policies before transactions are executed.

    Agents can spend money.
    They just can't prove they should.

    xBPP is the open standard that governs what AI agents are allowed to spend - before they spend it. Define policies once - spending limits, approved vendors, escalation rules - and every agent follows them. No custom code. No prayer. Just policy.

    Open standard
    Payment-rail agnostic
    Chain agnostic

    Who are you?

    Scroll

    THE AGENTIC GAP

    We solved Capability - the ability to act.
    We haven't solved Accountability - proving the action was safe.

    Capability without constraint is dangerous:

    An agent with access to a credit card could drain it in seconds

    A compromised agent could send funds to attackers

    Even well-intentioned agents make mistakes

    Currently, there are only two ways to control an agent:

    Prompt Engineering

    "Please don't spend more than $100"
    FragileNon-deterministicInjection attacks

    Hard-Coded Logic

    if (amount > 100) { reject(); }
    UnscalableRigid silosBreaks across jurisdictions

    Traditional solutions don't work:

    Manual approval destroys the value of automation
    Fixed limits are too blunt ($100 blocks both legitimate and fraud)
    No limits = unacceptable risk

    This creates

    The Agentic Gap

    The massive divide between what agents can do and what businesses dare let them do.

    THE AGENTIC STALEMATE

    STUCK IN PILOT PURGATORY

    "Thousands of high-capability agents sitting on local laptops, forbidden from touching real money because the liability risk is infinite."

    Market Reality 2026
    HOW IT ALL CONNECTS

    TRUST IS THE PRODUCT.

    A standardized middleware layer between Agent and World

    YOUR APPLICATION

    Agent wants to pay $50 to 0xABC...

    XBPP INTERPRETER

    Action + Policy → Verdict

    RESULT

    ALLOW | BLOCK | ESCALATE

    Step 1

    Intent

    Agent generates an action request

    "Pay $50 USDC to 0xABC on Base"

    Step 2

    Interpret

    xBPP checks against active Policy

    Constraints evaluated in real-time

    Verdict

    ALLOW

    Proceed with the payment

    BLOCK

    Stop; this violates policy

    ESCALATE

    Ask a human to approve

    Key Concepts

    ConceptWhat It IsExample
    ActionA proposed payment to evaluate"Pay $50 USDC to 0xABC"
    PolicyRules defining what's allowed"Max $100/tx, $1000/day"
    VerdictThe evaluation resultALLOW, BLOCK, or ESCALATE
    StateRunning totals and history"$340 spent today"
    PostureDefault risk toleranceAGGRESSIVE, BALANCED, CAUTIOUS

    Graduated Autonomy

    Trust builds over time - automatically

    Cautious Posture

    $10 limit, full human oversight

    Aggressive Posture

    $10k limit, autonomous execution

    Graduate between postures as trust builds - without rewriting a single line of code.

    Ready to test?

    Validate your agent policies in real-time.

    HOW XBPP CHANGES EVERYTHING

    Current State
    With xBPP
    Control

    Hard-coded if (amount > 100) inside the bot

    External Policy file: max_single: 100

    Update

    Redeploy the entire agent to change a limit

    Update the Policy; agent obeys instantly

    Audit

    "Check the logs and hope"

    Cryptographic Verdict attached to every transaction

    Liability

    Blame the developer

    Audit the verdict - the Protocol enforced the Policy

    XBPP: EXECUTION BOUNDARY
    PERMISSION PROTOCOL

    One question, answered before every agent action:

    The core question

    "Should this agent be allowed to spend this money?"

    xBPP is a policy engine for autonomous agents. You define the boundaries. The protocol enforces them.

    Declarative Policies

    Define behavior in external policy files, not buried in code.

    Open Standard

    Interoperable across agents, platforms, and jurisdictions.

    Transparent Verdicts

    Every decision is deterministic, auditable, and provable.

    What xBPP Does NOT Do

    Execute payments

    That's x402's job — xBPP governs, x402 executes

    Custody funds

    xBPP is policy, not a wallet

    Guarantee counterparty behavior

    Can't prevent a vendor from being malicious

    Replace legal contracts

    xBPP is technical enforcement, not legal agreement

    Where xBPP comes alive

    Vanar xBPP is a theatre of consequences.

    Watch identical agents, identical worlds - with one rule changed. See exactly where behavior diverges. Understand what policies actually do.

    SEE XBPP IN ACTION

    Real scenarios. Real risks. Real protection.

    Without xBPP

    Travel Assistant

    Risk: Over-booking premium flights

    Agent books $8,000 business class "because availability was low"

    With xBPP

    Travel Assistant

    Protection: Policy-enforced limits

    Policy caps single bookings at $800, escalates anything higher

    What Vanar xBPP Is NOT

    tell you what to do
    optimize your agent
    rank policies
    predict outcomes
    replace judgment

    It shows you what your rules mean.

    The decision is still yours.

    IN THE NEAR FUTURE, NO AUTONOMOUS AGENT
    WILL EXECUTE WITHOUT A SIGNED VERDICT.

    Market Context: The Agentic Economy

    Crossmint

    Building wallets for agents

    Kite AI

    Building the Agentic Internet layer

    Visa

    Researching Trusted Agent Registries

    Coinbase

    Exploring agent authentication standards

    Emerging platforms are standardizing infrastructure. They're all waiting for the same missing piece: a shared standard for agent governance.

    SSL

    Standard for moving data

    xBPP

    Standard for moving value

    xBPP transforms the "Wild West" of on-chain agents into a

    reliable, insurable, and scalable economy.

    EVERYTHING YOU NEED
    TO KNOW

    xBPP (Execution Boundary Permission Protocol) is the open standard for governing autonomous AI agent payments. It gives agents a declarative policy — what they can spend, with whom, and under what conditions — producing deterministic, auditable verdicts for every transaction.

    xBPP stands for Execution Boundary Permission Protocol. The “x” signals extensibility across payment rails. The “Execution Boundary” is the decision point where an agent’s spending intent meets policy enforcement before money moves.

    xBPP and x402 are complementary layers. x402 handles payment execution — moving money between parties using HTTP 402 flows. xBPP sits above it as the governance layer, deciding whether a payment should be allowed, blocked, or escalated before x402 executes it. Think of x402 as the engine and xBPP as the steering wheel.

    A policy is a declarative JSON config that defines an agent’s spending rules: per-transaction limits, daily/monthly budgets, approved counterparties, currency restrictions, time windows, and human escalation thresholds. Policies are evaluated deterministically — same input always yields the same verdict.

    Install @vanar/xbpp, create a policy with your limits (max_single, max_daily, require_human_above), and wrap your x402 client. Every transaction is checked before execution. Setup takes under 60 seconds.

    x402 answers “how do I pay?” — it is a payment execution protocol. xBPP answers “should I be allowed to pay?” — it is a governance protocol. They work together: xBPP evaluates the policy, and if approved, x402 executes the payment.

    xBPP is chain-agnostic by design. The reference implementation is built by Vanar and optimised for Base, but the protocol works across any EVM-compatible chain and is designed to be extended to any payment rail.

    Yes. xBPP is an open standard with an open-source reference implementation. The specification, SDK, policy templates, and test suite are all freely available and open to contribution.

    The behavior is the standard.

    BUILD AGENTS WORTH TRUSTING.