Each sector scenario page shows injection payloads written for its sector: a medical portal, a bank, a developer tool. This page shows the traffic the registry assigns to each sector. It does not show whether clients behave differently by sector: a callback is recorded on the canary URL, which carries no sector, so no callback can be credited to the page that planted it.
Hypothesis
Under test: a client that is served a scenario page completes an instruction-only step more often when the page's sector looks authoritative (healthcare, government) than when it does not. The seven pages share one domain, so whether a domain suffix such as .org changes the rate is not tested here. If the hypothesis holds, authoritative-looking pages receive less scrutiny from clients where they warrant more.
Status: this page cannot test it. A test needs each callback linked to the page that planted its canary, and callbacks attributed to agents rather than to crawlers and browsers. The current instrument records neither.
Loading sector traffic from the registry.
Sector scenarios
Explore realistic attack scenarios for each sector. These pages show how injection payloads are tailored to sector-specific contexts.
successRate is withdrawn: a callback is recorded on the canary URL, which carries no sector path, so every callback lands in the general bucket and none can reach a named sector. A per-sector rate would divide outcomes that cannot be attributed to a sector by rows that can. Per technique, follow-through rows carry no technique tag, and a legacy-tagged callback counts inside its own denominator, so the same division fails there. Re-deriving the ratio from the published counts reproduces the same invalid division. A rate returns when a callback can be linked to the page that planted its canary.
Requests whose URL matches no sector scenario page: agentpwn.com's own pages and every callback, because the canary URL carries no sector. This is why per-sector callback counts cannot be read from this breakdown.