Your AI can write a memo. Can it issue a refund? Safely?

Most agents can only read and retrieve. Bounded lets them act — issuing refunds, updating records, moving work forward — without handing over a raw API. You define each action with hard limits, an approval step, and a full audit trail. The agent does exactly that, and nothing more.

Coming soon Talk to us

We're onboarding early customers by hand. Talk to us for access.

Claude
Copilot
Bounded
limits · approval · audit
Payments
CRM

The contract

An action is a contract over an API.

You write the bound once. The agent sees only what you declared — no extra arguments, no second endpoint, no way around the limit.

refund.issue · bound
action: refund.issue
arguments:
  order_id:  { type: string }
  amount_usd: { type: number }
rules:
  - cap: amount_usd <= 500        # hard limit — never exceeded
  - approval: amount_usd > 200    # ask a human first
downstream:
  method: POST
  url: https://api.acme.com/orders/{order_id}/refund

The proof

Every attempt is recorded.

Allowed, held for approval, or blocked — each call is tied to an identity and written to a hash-chained log you can verify. Nothing runs off the record.

Audit trail hash-chained
  • allowed
    refund.issue
    $140.00 · order_8821
  • pending
    refund.issue
    $420.00 · order_8830 · sent for approval
  • denied
    refund.issue
    $900.00 · over the $500 cap

entry_hash = sha256(prev_hash · canonical(entry)) · verifiable by walking the chain

Three steps to a governed action

Define the bound, connect your agent, and let it act. The limits hold on every call — you don't have to trust the agent to behave.

1

Define the bound

Declare the arguments, the hard limits, when a human must approve, and the downstream API. A draft is reviewed and activated — the editor can't approve their own change.

2

Connect the agent

Point Claude, Copilot, ChatGPT, Cursor, or your own agent at the workspace with a scoped bearer token. The agent discovers exactly the actions you granted it — and no others.

3

Act, within bounds

Every call runs the enforcement pipeline: arguments validated, limits checked, approval gated, then the downstream request fires. The result is recorded before the agent ever sees it.

Retrieval is solved. Action is the hard part.

An agent that reads your data is useful. An agent that changes your data is valuable — and dangerous. The gap between the two isn't a better model. It's governance: a way to grant write-level power that a security team can actually approve.

Bounded is that layer. Capability is the hook; containment is the proof. You give the agent real power, and you can prove it can only use the part you authorized.

Hard limits

Capped at $500. At $501 it stops. Limits are enforced server-side, not suggested in a prompt.

Human approval

When a call crosses a threshold, it pauses and waits for a person to approve or reject.

Full audit trail

Every attempt — allowed, pending, or denied — is hash-chained and tied to an identity.

Separation of duties

The person who writes an action can't be the one who approves it. The log records both.

Give your agents real power. We make sure that's all they can use.

Define a bounded action, point an agent at it, and watch the limit hold. It takes a few minutes.

Coming soon Talk to us
bounded

The governed action layer for AI agents. Action, within bounds.

Sign in →