KNOW YOUR AGENT
Verify signed agents before you decide how to treat them
A display name or user-agent string is a claim, not proof. Know Your Agent verifies supported signed-agent identity claims and records the result beside the risk read.
WHAT YOU GET
Evidence about an automated client's identity claim
Verification shows possession of an accepted agent credential under a supported protocol. It does not show intent.
Verification result
Verified
A verified claim adds context.
Unsigned
Stays unknown, not hostile.
Invalid signature
An invalid signature is a risk signal.
Expired
Expired signatures are detected.
Replayed
Replayed signatures are detected.
Recorded beside the risk read
The result contributes to the risk read. Any allow, challenge or deny decision stays in your own logic.
A verified agent still passes your normal authentication and authorization.
Signed-request verification
Supported signed-agent requests are checked against public key references from the agent's declared signing authority. Expired and replayed signatures are detected.
A clear result state
Each request resolves to a plain state: verified, unsigned, invalid signature, expired, or replayed. Unsigned traffic stays unknown, not hostile.
Context beside the risk read
The result is recorded beside the risk read and contributes to it. A verified claim adds context. An invalid signature is a risk signal.
Operator view
Review agent identity claims and verification results in the Agentic Security view of the console. Record which verified agents your tenant treats as allowed, denied, or observed.
HOW IT WORKS
From a signed request to a recorded result
Know Your Agent evaluates claims on the traffic your domains receive. Your own logic decides what to do about them.
External agent
Signed request
A supported agent signs its request and declares where its public key can be found.
Noxtica
Verifies and records
Signed request
Checks the signature, key reference and validity window. The result is recorded beside the risk read.
Your policy
Your own logic
Result state
Observe matches and non-matches, review unknown agents and apply your own policy.
Noxtica
Enforcement
Enforcement is roadmap, not part of this offer
Roadmap
The agent signs its request
A supported agent signs its request and declares where its public key can be found. Agents that do not sign skip this path and stay unknown.
Noxtica verifies the claim
The signature, key reference, and validity window are checked. The result is a clear state, including unsigned, expired, and replayed.
You observe and decide
The result sits beside the risk read. Observe matches and non-matches, review unknown agents, and apply your own policy in your own logic.
WHERE IT FITS
Questions it helps answer
Use it when your application receives automated traffic and needs more than a name to go on.
Which agents identify themselves?
See which automated clients present a supported identity claim, and whether that claim verifies.
Is this claim known to us?
Separate a verified claim from a verified claim your team has not reviewed yet. Observe, then update your policy deliberately.
What should unknown traffic do?
Define a fallback for unsigned, invalid, and unsupported traffic. Observe it, or apply your existing bot-risk policy, until you understand its purpose.
HOW TEAMS CONNECT
Request verification and an operator view
Know Your Agent has no public sample code. Setup runs through your account team and depends on tenant provisioning.
- Request path: supported signed requests are verified where your domains receive traffic. This needs compatible signing and an edge integration.
- Server verification: your backend can ask Noxtica to verify an agent request with a server-held credential. Where provisioned, it can also carry verified agent context in an agent identity envelope.
- Operator console: review verification results in the Agentic Security view. Access follows operator role and tenant scope.
LIMITS TO PLAN FOR
What Know Your Agent does not do
Do not make it a launch dependency until your account team confirms eligibility and the agent identity path.
- Controlled availability. Eligibility, supported identity standards, and setup depend on tenant provisioning. Confirm with your account team before you plan around it.
- Verify and observe today. Enforcement is roadmap and is not part of what this page offers. Any block or challenge based on agent identity is your own logic. Nothing here commits scope or date.
- Verification is not trust. It does not prove intent, delegated authority, or the person behind an agent. A verified agent still passes your normal authentication and authorization.
- Results are evidence. Key rotation, authority changes, and malformed signatures change outcomes. Agents without a supported identity path stay unknown. Your team makes the decision. Read the disclaimer.
Scores, reasons and verdicts are probabilistic estimates, not statements of fact. Your team makes every decision.Read the disclaimer
FAQ
Common questions
Does Know Your Agent block unwanted agents?
Not today. It verifies and observes. Enforcement is roadmap, so apply any allow, challenge, or deny decision in your own logic.
Is it the same as MCP or the AI Assistant?
No. Those let your own agents or operators read approved Noxtica context. Know Your Agent evaluates agents that arrive as traffic.
What happens to agents that do not sign requests?
They stay unknown. Know Your Agent does not identify every bot. Keep a fallback policy based on browser, device, network, and route evidence.
Related products
KNOW YOUR AGENT
Confirm eligibility before you plan around it
Tell the team which agents reach your domains and how they identify themselves. We confirm provisioning and what verification can show you today.