Policy
Rules decide every agent action before it runs: amount caps, per-action allowlists, and segregation of duties — the agent that drafts an invoice can never be the one that approves it. No decision means deny, by construction.
Governance for AI agent actions
Agent Sentry — the governance gate for AI agents on your books
Agents act. Policy decides. Humans approve anything high-risk. Every step is evidenced in a tamper-evident ledger — so your AI can work your systems of record without anyone losing sleep.
Free to start with your Google Workspace account · talk to us
Evidence
A tamper-evident ledger records intent, decision, and outcome for every action, anchored in an independent transparency log and sealed into immutable, write-once storage. If the evidence can’t be written, the action doesn’t execute.
Approval
High-risk actions pause for a human. Approvers are notified with a secure link, see the exact transaction, and sign off with a fresh-MFA login — an agent credential can never satisfy an approval.
How it works
Your agent
Your AI assistant
proposes actions — create an invoice, record a payment, pay a bill.
Agent Sentry
The Sentry gate
evaluates policy, writes evidence, and routes anything risky to a human approver.
Your systems
Systems of record
only ever sees actions that passed policy — each one traceable to its evidence record.
Put a sentry between your agents and your ledger
Sign up with your Google Workspace account and connect your systems in minutes.