Features
Every capability of the AI Agent Firewall
A complete AI agent firewall, ready out of the box: verification and identity, voice and caller security, threat detection, data protection, policy and operations, and an API for your own apps.
Verification and identity
Before a sensitive tool runs, the person proves who they are with a daily word, a passphrase, an authenticator code or a phone approval, and a duress word raises a silent alarm.
Explore verification and identity →Word of the day
A short common word that changes at midnight, visible only to the members you choose, which a caller says inside a random three-word phrase so a replayed recording fails.
Duress word
A member under pressure says the duress word instead of today’s word: the agent responds as if verification passed, every sensitive tool quietly says it is unavailable, and admins are alerted.
Verify before use (tool authorization)
Turn on Verify before use on any tool and the agent pauses to ask for the word of the day, a passphrase, a TOTP code or a push approval, one or all of them, before the tool runs.
Voice and caller security
Every caller utterance is scored for synthetic speech, every inbound number is checked against the carrier’s caller ID result, and agent voices carry a watermark that exposes replays.
Explore voice and caller security →Voice deepfake detection
Every thing a caller says on a phone call or in voice chat gets a synthetic-voice score from 0 to 1, labeled Real, Synthetic or Unsure, and a policy action applies at 0.5 or above.
Caller ID check (STIR/SHAKEN)
On every inbound call the firewall reads the carrier’s STIR/SHAKEN result and applies your action for Verified, Partly verified, Not checked and Failed numbers.
Voice watermark
Autessa agent voices carry an inaudible watermark, so replayed agent audio is treated as synthetic, and the voice score API reports whether a clip came from an Autessa agent.
Threat detection
The firewall flags prompt injection in messages and tool results, learns each person’s normal requests to spot anomalies, and reviews conversations for pressure tactics.
Explore threat detection →Prompt injection protection
The firewall scores every user message and every tool result for prompt injection, flags it, and requires approval for any later tool call in the same run that sends data to another site.
Unusual-request alerts
The firewall learns each person’s usual tools, channel, hours, location and volume on each agent, and raises a plain-language alert when a request stands out.
Pressure-tactics check
Before a sensitive tool runs, an AI review checks the recent conversation for social-engineering cues such as urgency, impersonation, gift cards or bank changes, and quotes them.
Data protection
Cap the sensitive records anyone can extract, allowlist the tools and sites an agent may use, and block secrets or personal data from leaving.
Explore data protection →Sensitive-record limits
Cap the sensitive records one person can receive per conversation and per rolling 24 hours, counting only tools and AutessaDB tables your team has marked as sensitive.
Tool and data controls
Allowlist the tools an agent may call and the sites it may send data to, hold new email domains for approval, block secrets leaving, cap rows read and stop tool loops.
Policy and operations
Start in Monitor, write rules and exceptions you can preview against real traffic, approve held requests, watch the dashboard and red-team agents with the simulator.
Explore 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.
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.
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.
Developers
Your own chat, agent and voice applications call one REST API for every phase of a turn and get back an action, reasons and findings.
Explore developers →One firewall for every agent and every channel.
Chat, text, phone and your own apps, under one policy you can watch before you enforce.
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