No traffic proxy
Volance does not proxy or serve your traffic. Your server calls the API and applies its own policy — so if you wait for a result before responding, that call adds latency. Configure timeouts and a fallback that fit your application.
BOT & AGENT DETECTION, EXPLAINED
Volance is an explainable bot and agent detection API. Add one script to your pages and one API call from your server, and each session gets a 0–100 human-likeness score, a human, agent, or bot label, and the signals behind the verdict. Your server decides whether to allow, flag, or block. Verified AI agents can pay for access, and Volance takes no cut.
Free to start · no credit card. One script on your page. One API call from your server.
Varied interaction timing and consistent client signals.
A score is an estimate. Not proof of identity.
Human sample: score 96 of 100, verdict human, action allow.
WHY VOLANCE
Volance scores the session, explains the evidence, and leaves enforcement on your server.
Volance does not proxy or serve your traffic. Your server calls the API and applies its own policy — so if you wait for a result before responding, that call adds latency. Configure timeouts and a fallback that fit your application.
Not a black box with an opinion. Each response names the checks that ran, what they read, and which ones moved the score — so a decision can be explained to your team, your customer, or an auditor.
Automation detection and access policy are different questions. Volance separates human, agent and bot, and lets verified agents through by policy — because the web is shared with automation that belongs.
On-site signals only. No cross-site identifiers, no fingerprinting unless you ask for it, hashed IP addresses, and raw scores purged after 30 days. Boundaries you can put in writing.
THE WEB HAS CHANGED
Volance separates humans, agents, and bots, so legitimate automation can be allowed instead of blocked. A binary bot check misses the bigger picture. Volance gives you a human-likeness score, a classification, and the signals behind the verdict—so you can make a more informed call.
Integration overviewEnable on-site behavioral and client signals where you need them. Collection is an explicit integration choice, not a hidden default.
Inspect the score, classification, and contributing evidence—not just a yes or no. See why a request deserves a closer look.
Keep enforcement on your server. Allow, flag, or block according to your application’s policy. Legitimate automation can belong.
A RESPONSE YOU CAN REASON ABOUT
The API returns a score, a classification, a verdict, and the evidence behind it — a readable response for the application you’re already building. Keep secrets and enforcement on your server.
Illustrative excerpt. Not a live response.
Read the API reference{
"score": 96,
"verdict": "human",
"action": "allow",
"classification": {
"label": "human"
}
}Changes with the sample inspector. No API call is made.
THE PRIVACY DESIGN
Behavioral detection needs clear boundaries. Volance collects on-site only, keeps no cross-site identifiers, hashes IP addresses, and purges raw scores after 30 days.
How Volance handles privacyFOR THE AGENTIC WEB
Verified agents pay inline with HTTP 402 and x402, then your edge serves the content. Volance decides who is asking and what they should pay, then hands your edge a ready 402. You set the price; in x402 mode your edge settles to your wallet through your facilitator. Preview — live production settlement is not generally available.
Human, agent, or bot — plus whether the agent's Web Bot Auth claim actually verifies.
Rules by path and caller. Charge verified agents, allow humans free, block abuse.
Your edge enforces the 402; in x402 mode funds settle to your own wallet through your facilitator. Volance never holds them.
MAKE THE NEXT REQUEST MAKE SENSE.
One script on your site.
One API call from your server.
Every signal behind the verdict, in your own portal.