Policy and operations

Alerts and approvals

Every AI Agent Firewall decision lands on one sortable alerts page with its score, cues and a link to the conversation, and tool calls held for approval get Approve and Reject buttons, also in each admin’s inbox.

Turn on the firewall

Alertssorted by timeSAMPLE

Payroll Assistant · get_payroll · synthetic voice 0.9114:06 · enforcedBLOCK

Vendor Desk · update_bank · pressure tactics13:58 · Approve / RejectREQUIRE_APPROVAL

Support · search_customers · daily limit13:41 · enforcedBLOCK

HR Assistant · new country13:20 · recordedFLAG

The problem

Security signals are only useful if someone can act on them quickly, with enough context to decide.

How alerts and approvals works

STEP 1

Everything is recorded

When, which agent and tool, the check, the action, and whether it was applied (Enforce) or only recorded (Monitor).

STEP 2

Open the details

The voice score, what was unusual, quoted pressure cues, with a link to the conversation or call.

STEP 3

Approve or reject

Approving lets that agent run that tool with that exact request. The security notification in the inbox has the same buttons.

Every setting

Sorting

Every column can be sorted.

Sampling

Requests that pass every check are recorded as a sample, so the list stays readable.

Retention

Alerts, scores and verification records are deleted after 30 days.

API

List alerts through the API for your own SIEM or tooling.

Who can do what

Admins

View alerts and approve or reject held requests.

What to know

An approval applies to that exact request only, not to the tool in general.

More in Policy and operations

Custom rules and exceptions →Ordered rules add an action and exceptions remove findings, using conditions on channel, tool, identity, voice score, country, time and text, previewed against the last 7 days.Monitor and Enforce →Monitor records what every check would do without changing what customers see; Enforce applies the actions. Each agent uses the company policy, its own custom policy, or none.Monitoring dashboard →One dashboard shows checks, catches, blocks, verifications and points for every agent and API policy, with trends, what was caught, top rules and top identities.Attack simulator →A dry-run red team: built-in prompt injection, pressure, passphrase guessing, word replay, anomaly, exfiltration and cloned-voice attacks against an agent, with a scorecard of what got through.All of Policy and operations →

← Monitor and Enforce Monitoring dashboard →

FAQ

Alerts and approvals: questions

Can a human approve an AI agent’s action before it runs?

Yes. Any check can be set to Require approval; the tool call waits and admins approve or reject from the alert or their inbox. On your instructions, an Autessa agent can also call or email the admin on duty so a held request does not sit waiting; the decision itself stays in the app.

Turn on alerts and approvals.

Every decision lands on a sortable alerts page with the score, cues and a link to the conversation; held tool calls get Approve and Reject, also from the admin’s inbox. Start in Monitor and see what it would catch on your real traffic.

Turn on the firewall

Starts in Monitor · nothing changes for your customers until you enforce

100% customizable, code and all

Make Autessa AI Agent Firewall fully your own.

Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.

Start free and make it yours