00 — THE THRESHOLD
2026 / Q3

AI agents are already shopping your store. Give them rules. Keep the proof.

Sill verifies every AI agent, enforces your store rules at checkout, and keeps a tamper-evident record of every visit, cart, and purchase.

https://
Sill checks your store's agent readiness in seconds and shows what agents could do on your platform.
FREE DISCOVERY NO CARD REQUIRED SHOPIFY · WOOCOMMERCE · ANY STORE
01a — RUNTIME · AUTOMATEDILLUSTRATIVE DECISION PATH
VISITOR ARRIVES
STOREFRONT
mill-valley-coffee.com
MANDATE PRESENTED
awaiting…
POLICY EVALUATION
signature_valid
rate_limit · 60/min
sku_in_allowlist
amount_under_cap
01b — RUNTIME · HUMAN IN THE LOOPESCALATION PATH FOR AMBIGUOUS MANDATES

Set a dollar threshold. Above it, your team reviews and approves before the agent's order goes through.

High-value agent requests escalate to a named reviewer on your team. Approve or reject; either way the decision is written into a signed audit record with the reviewer's stamp. Simulated below.
WESTPORT JEWELRY POLICY · CAP $10,000 · HUMAN REVIEW ABOVE $1,000

HIGH-VALUE MANDATE
STOREFRONT
westport-jewelry.com
POLICY · WITH ESCALATION(5 of 29 rules shown · illustrative)
signature_valid
rate_limit · 60/min
sku_in_allowlist
amount_under_cap
high_value_review_required
OPERATOR REVIEW
idle…
ESCALATION → REVIEW → SIGNED RECORD·see how policy + audit work →
02 — AUDIT TRAILEVERY MANDATE · SIGNED · EXPORTABLE

Every mandate becomes a signed, timestamped record.

Every mandate leaves a signed, tamper-evident record — what was authorized, how it was resolved, and what settled — verifiable by anyone with standard open crypto. No Sill software required.

INTENTaction · merchant · SKU · amount cap · currency
SETTLEoutcome record, same format regardless of processor, linked to the decision
HUMAN REVIEWpseudonymous reviewer id + decision time
CHAINed25519 · Merkle-chained · append-only
VERIFYany third party · in-browser at sill.so/verify
FRAMEWORK MAPPINGS · OWASP · MITRE ATLAS · NIST AI RMF
Sill
AGENT GOVERNANCE INFRASTRUCTURE
DOCUMENT ID
SB-20260711-PM_8A3F9C12

ISSUED
7/11/2026, 9:15 AM
MANDATE AUDIT BUNDLE
An evidentiary record of one agentic transaction.
Captures the cryptographic identity, signed policy decisions, and Merkle-chained audit record for mandate pm_8a3f9c12. For internal compliance, external auditors, and regulators.
Mandate verified.
All policy checks passed. Signed audit record written.
01Mandate parties
MANDATE IDpm_8a3f9c12
ISSUING AGENTChatGPT
END USER[email protected]
USER INTENTBuy 2× Espresso Roast under $50
AMOUNT CAP$47.20 USD
SIG ALGORITHMed25519
+ 4 MORE SECTIONS · DECISION TRACE · MAPPINGS · ANCHOR · ATTESTATION

How Sill governs every agent interaction: identity, intent, proof

03 — IDENTITY, INTENT, PROOF

IDENTITY

Every request to the merchant carries a signed agent card naming who the visitor is, who deployed them, and what their public key proves. Unsigned traffic gets handled by your existing fraud rules. Signed traffic enters a different flow.

INTENT

Before money moves, the agent presents a signed mandate: the SKU, the cap, the merchant, the expiry, the signature. Sill evaluates it against your policy. A mandate that fails any rule never reaches Stripe to authorize the charge. Sill evaluates and signs; it never custodies funds.

PROOF

You get exportable proof of exactly what an agent was allowed to buy — action, merchant, SKU, amount cap, currency — and that no one changed it after the fact. Every evaluated mandate writes an append-only, signed record; where settlement follows, a matching signed record binds the outcome back to the authorizing decision. Any third party can verify the crypto with our published key.

04 — GUARDRAILSCATEGORIES · NOT THE RULE LIBRARY

What Sill enforces, before money moves.

Each mandate is evaluated against six categories of policy. The categories are public; your site's specific rules and thresholds live behind authentication.

Defaults ship safe. Customization is opt-in. The DSL is documented, versioned, and testable.

IDENTITY
Who is the visitor

Signature validity, agent card freshness, organizational provenance, key rotation. Unverified agents are denied at the threshold.

RATE & VOLUME
How often, how fast

Per-agent, per-merchant, per-time-window limits. Burst windows. Anomalous concurrency. Tunable per skill.

TRANSACTIONAL
What and how much

Amount caps, SKU allowlists, currency restrictions, geographic and shipping rules, bundle constraints.

BEHAVIORAL
Pattern over time

Sequence analysis, velocity changes, suspicious ordering patterns, anomaly detection across an agent's request history on your site.

CUSTOM POLICY
Your rules, your code

Merchant-defined policy expressions in a constrained DSL. Versioned, testable, deployable from the dashboard or via API.

AUDIT
Proof and retention

Every decision is logged, signed, and retained. Exportable as a server-rendered HTML audit bundle.

RED TEAM SIMULATORATTACK YOUR OWN POLICY BEFORE SOMEONE ELSE DOES

Test your active policy against documented attacks.

Sill's policy is a 29-rule engine mapped to MITRE ATLAS, the OWASP LLM Top 10, and the OWASP Top 10 for Agentic Applications. Each rule is a named, versioned guardrail — and every evaluation is recorded in a signed, Merkle-chained audit record.

You see exactly which scenarios your policy catches and which slip through — before they become an incident.

POLICY29 rules across 6 categories
SOURCESMITRE ATLAS · OWASP · AP2
SCOPEDPer rule or globally
RECORDEDOutcome logged with policy version
EXAMPLE: POLICY EVALUATION
0 CAUGHT0 MISSED0/6
AWAITING…
AWAITING…
AWAITING…
AWAITING…
AWAITING…
AWAITING…
POLICY ENGINE · 29 RULES · 6 CATEGORIES — MAPPED TO MITRE ATLAS / OWASPPAUSED
05 — WHERE SILL SITS

One script tag, between the agent and your stack.

Sill runs at the edge, in front of your existing commerce stack. Agents arrive with signed HTTP requests; Sill identifies them, evaluates each request against your policy, and forwards approved actions to your existing processor — it never custodies funds.

INSTALL TIMEone script tag
SIGNATURESed25519, independently verifiable
AUDITsigned, exportable, independently verifiable
How Sill sits between an AI agent and your payment stackAI Agentsigned HTTPrequestarriveSILLEdge LayerVerify identityEvaluate policyAuthorize or rejectSign audit recordcdn.sill.so/embed.jsapprovedStripe Connectpayment rail
AGENT-READY SURFACES

Signed surfaces an arriving agent can read — and anyone can verify.

When a site is verified, Sill publishes machine-readable endpoints at the edge so agents can discover its identity, skills, and catalog. Each is signed with an ed25519 envelope, and the signing key is public — so anyone can verify a signature without trusting Sill.

A2A agent carda2a / 1.2 · signed
GET edge.sill.so/v1/agent-card/{site_key}.json

A signed, A2A-compatible card per verified site. Advertised skills are filtered to what the site actually exposes — no overclaim.

MCP serverstreamable-http · live
POST edge.sill.so/v1/mcp/{site_key}

A live MCP endpoint exposing the site’s backed skills as tools. tools/call invocations are recorded to the audit log.

ARD catalogard / 1.0 · signed
GET edge.sill.so/v1/catalog/{site_key}.json

A signed ai-catalog.json. Each trust manifest carries a detached ed25519 JWS over its JCS-canonical payload, attesting identity and provenance; host identity is did:web.

Public JWKSed25519 · EdDSA
GET edge.sill.so/.well-known/jwks.json

The signing key, published openly. Anyone can verify the card and catalog signatures with standard JWS + JCS tooling — no Sill code required.

06 — STANDARDS
FRAMEWORK MAPPINGS · NOT CERTIFICATIONS

Sill maintains public, control-by-control mappings between its guardrails and the security frameworks below.

These are implementation mappings, not certifications or attestations. Merchants remain responsible for their own regulatory compliance and any accredited audit their regulators require.

Pricing — agent commerce governance, priced like software

07 — PRICING
DISCOVERY AVAILABLE · PRICING IN DEVELOPMENT
DISCOVERY
$0during beta
LIVE
Agent identity logging
Your store published to AI agent discovery services
Third-party verifiable signed identity
No payment authorization
Add website
TRANSACTIONAL
In validationpricing in progress
LIVE STRIPE RAIL · IN VALIDATION
Everything in Discovery
Signed mandates + policy engine
Stripe payment authorization
Human-in-the-loop escalation
Signed, Merkle-chained audit records
Talk to us
ENTERPRISE
Customvolume & SLA
CONTACT
Everything in Transactional
Contact us to discuss requirements
Contact sales
WHAT IS A MANDATE?

A mandate is a signed agent request to perform a sensitive action — a checkout, refund, order lookup, or shipment update. Each mandate is evaluated against your policy and produces one audit record. When paid tiers launch, billing is per mandate evaluated, not per agent visit.

08 — FAQ
OBJECTIONS · ANSWERED

Sill touches payment authorization, agent identity, and audit records — the parts of your stack where unclear answers matter.

What can I do with the free Discovery plan?

Identify agent traffic in your logs, publish a signed skill manifest so well-behaved agents discover your store, and see which actions agents would request. No payment authorization, no transactional authority — useful immediately for visibility.

How long does it take to install?

Under five minutes. Enter your domain, verify ownership, and add a one-line script tag from your dashboard. Your signed agent card and AI catalog go live immediately.

Can agents complete checkout automatically without my approval?

Only if you allow it. Each merchant defines what agents can do unattended (typically small repeat purchases) and what requires human review (high-value, refunds, account changes). The default policy ships conservative.

Does Sill process payments?

No. Sill evaluates and signs mandates; payment authorization stays with your existing processor (Stripe). We never touch funds.

Does this replace Stripe, Shopify, or WooCommerce?

No. Sill sits in front of your existing commerce stack. Approved actions are forwarded to your checkout, refund, or order systems unchanged. You keep the same processor, the same platform, and the same data ownership.

Do I need to support new agent protocols (A2A, AP2, MCP) yourself?

No. Sill normalizes inbound agent requests and presents them to your backend in a single format. As new protocols stabilize, we add support; your integration stays the same.

Can I block unknown or unverified agents?

Yes. The default policy declines mandates from unsigned or unverified agents. You can allowlist specific agent identities, require minimum verification levels, and rate-limit by principal.

Does Sill expose my private rules?

No. The rule categories are public so visitors and auditors understand what Sill enforces. The specific thresholds, allowlists, and policy logic for your site live behind authentication and are never returned in error responses or visible to agents.

Will this work with Shopify, WooCommerce, or a custom store?

The embed script installs on any site with one line of HTML. The Stripe transactional rail is live in a controlled single-account validation; broader onboarding is founder-gated pending compliance review. Shopify and WooCommerce connectors are live in test mode; live-mode settlement is not yet enabled.

Different question? Email [email protected] and we'll answer it directly.

09 — START

Start seeing which AI agents are visiting your store. Free, takes minutes.

Every registered store gets published to AI agent discovery services with a signed identity that any third party can verify. Works with Shopify, WooCommerce, or any platform.