Identity
Checks who the agent is, who it acts for, and how long its authority holds.
- OWASP Agentic ASI03
- OWASP Agentic ASI10
- OWASP Agentic ASI07
- OWASP LLM06:2026
- ATLAS AML.T0091.001
Every agent decision on your site is signed. The key is public. The verifier runs in your browser.
Want to block attackers and refuse attack paths? That lives on Security.
For every agent action we answer three questions, in order. Who is this? What are they allowed to do? What happened?
Who is at the door. Known agents are recognized by their signed card. Unknown callers are still logged, never silently waved through.
What they want to do. A signed request is checked against your own rules — spend caps, geography, human review — before anything settles.
What happened. The outcome is signed and linked to the one before it, so you can show what your site decided, long after the fact.
A request arrives. Rules evaluate it. Some actions wait for a person. The outcome is signed either way.
An agent calls one of your skills.
Your active policy — spend caps, geography, human-review threshold — is checked before anything settles.
Over your threshold, the action waits for a person. A reviewer approves or declines from the dashboard.
Approved, held, declined — each outcome gets a signed record, linked to the one before it.
Download a bundle from your dashboard. Open the verifier. The bundle never leaves your browser; nothing is sent to Sill.
Every agent action runs through your rules. The six categories are public; your thresholds live behind sign-in.
An allowlist of registered agents, and signed agent-to-agent handoff rules where delegation is used. Rules apply to actions; page visits are recorded, not blocked.
A per-transaction spend cap, a daily spend cap per customer, and a budget cap across agents from the same customer organization.
Per-agent, per-IP and site-wide limits on actions and tool calls, plus an MCP discovery rate limit at the edge.
Refunds and cancellations over a threshold wait for a person. Recurring charges need their own user-signed authorization. A reviewer approves or declines from the dashboard.
Block instruction-override text, block invisible-character attacks, cart stays under the customer’s budget, cart currency matches the authorization, short-lived order authorizations and order payload size limit, no fake-urgency orders and no drip pricing, block orders that contain secrets, and an emergency kill switch.
Country allow / deny for agent actions.
You attack your own rules with the same library a real agent might try. Nothing is sent, nothing is charged. Every miss comes with the fix.
Press Run to attack the sample policy.
Twenty-two attacks will stream in, grouped by category. Two land as misses so you can see the fix.
One card per attack category Sill tests. The ids are from public registers — the OWASP agentic and LLM Top 10s, MITRE ATLAS, and CWE — so a reviewer can trace what we test back to the catalogue they already use.
Checks who the agent is, who it acts for, and how long its authority holds.
Verifies the request is cryptographically signed, in-window, from a registered agent, and not replayed.
Keeps a transaction inside the amount, currency and ceiling the user signed off on.
Catches instruction overrides, smuggled characters, leaked secrets and dark patterns in the request fields.
Keeps actions inside the signed user, the approved skill list, and sensible body limits.
Your existing processor holds the card, authorizes the charge, settles to your account and pays you out. Sill issues the signed authorization and the signed record. No funds enter a Sill-controlled account at any step; raw card numbers never enter any Sill system.
We say what we have shipped and nothing more.
Type your domain. See your site the way an AI assistant reads it.
Type your domain. See your site the way an AI assistant reads it.
Type your domain. See your site the way an AI assistant reads it.