Build your own ClickCease alternative with your AI agent
Your own invalid-traffic detection, running in your warehouse next to your conversions: every request, session and conversion on your site gets a verdict with the evidence behind it, built to replace click-fraud tools such as ClickCease and traffic verification contracts whose rules you cannot inspect.
Paste one prompt. Your agent builds your version.
Paste it into Claude Code, Codex or Cursor. Your agent reads the blueprint, asks how your company works and shows you a one-page plan. Once you approve it, the agent builds your version on SegmentStream, turns every invariant into a test and deploys when you say so.
Read https://github.com/segmentstream/build-your-own/blob/main/AGENTS.md and follow it.
Build our Ad fraud and bot detection from the ad-fraud-detection blueprint.What your agent builds
A verdict with its evidence for every website session and conversion: the full rule set SegmentStream runs in production for bots, spoofed crawlers, automation, data-centre traffic and AI agents.
What it is not
- Not real-time blocking. It judges traffic after the fact, in shadow mode; blocking belongs to your CDN or firewall.
- No cross-customer reputation: your build sees your own traffic only.
- Not a guarantee. A verdict is evidence for your own decisions, such as a refund claim to an ad platform; review it before you act on it.
- Rules only. It does not check Web Bot Auth signatures cryptographically, use reverse DNS, or report phantom clicks yet; VARIATIONS.md describes how.
- The full rule set SegmentStream runs in production, as of this version: 17 hard rules in a fixed precedence, 9 weighted weak rules (one reserved), every threshold, and the class, verdict and GIVT/SIVT logic, recalculated over all your history on each run.
- Every session classed as human, likely human, unverified, AI assistant, autonomous agent, crawler, spoofed, script or internal, with a verdict (valid, suspicious, invalid) and the rules that decided it.
- The evidence for any session or lead: each rule that fired, on which request, with the values it saw (user agent against client hints, TLS fingerprint, Fetch Metadata, network, script runs, per-address volume).
- Claims checked against 21 public reference lists: a crawler or AI agent from an address its vendor does not publish is caught as spoofed; cloud, hosting and private relay ranges are told apart.
- Form submissions judged with their own request and their session, so detected bots do not count as leads and real people behind a VPN still can.
- Invalid traffic by reason, class, channel, platform, network and country, and the AI agents and crawlers that read your pages, verified.
The hard parts are already running.
Your agent writes only the business logic. SegmentStream runs the infrastructure under it, so nobody on your team builds or operates it.
- Real-time collection of page views and server request evidence through one endpoint and a first-party tracker, into warehouse tables in your region (set up by the marketing analytics blueprint).
- Scheduled ingestion of public reference lists (AI agents, crawlers, cloud and proxy networks) through public, read-only connections that need no credentials.
- Managed pipelines that classify every session on a schedule (hourly in our own production), with backfills, retries and checks.
- The optional Google Ads connection, with credentials kept by SegmentStream.
- ClickCease
- CHEQ
- TrafficGuard
- Spider AF
- Fraud Blocker
- ClickGUARD
- HUMAN Security
- DoubleVerify
- Integral Ad Science
- Anura
- Polygraph
Product names are trademarks of their owners, named only to describe what the blueprint can be compared with.
What’s in the blueprint
Five files your agent reads before it builds anything. Each opens on GitHub.
- BLUEPRINT.mdThe designRecords, actions, roles, workflows, data flow, screens and the build order.
- INTERVIEW.mdThe interviewWhat your agent asks about your company, and what each answer changes.
- INVARIANTS.mdThe invariantsRules that must hold in every version, each one a test in your project.
- VARIATIONS.mdThe variationsWhere companies legitimately differ, and the default for each.
- PITFALLS.mdThe pitfallsMistakes agents already made in production, so yours does not.
Blueprints are provided as is; see the disclaimer.
Related solutions
Build your company’s own software. Let agents run it.
$ segmentstream workspace create acme --region eu --subdomain acme
▸ Created workspace acme · database, warehouse and a private git repository
$ segmentstream workspace WORKSPACE-ID clone acme && cd acme
▸ your coding agent builds what your team asked for and runs the checks
$ git add -A && git commit -m "First version"
$ segmentstream deploy
▸ app, company agent, API and MCP tools live for your invited members
Our engineers build the first version on your data in two weeks, at a fixed price.
Or build it yourself with your coding agent:
curl -fsSL https://segmentstream.com/install.sh | sh