The cross-harness, self-improving memory layer for AI agents. Join our community: https://discord.com/invite/zdNChS2fBu
Go agent-ai-cliagent-harnessesagent-securitybrowser-extensionciclaude-codecloudcodexcursordetection-engineeringjevknowledge-basememorymobile-device-managementobservabilitysecuritysecurity-information-and-event-managementsecurity-toolstelemetrytraces
test(windows): gate user-mode capture, now that a run has demonstrated it
shukieshah50f1ecdc37c0
diff --git a/beacon-sandbox/scenarios/w02-install-supervised.yaml b/beacon-sandbox/scenarios/w02-install-supervised.yamlindex 328b3f29..69a461ff 100644--- a/beacon-sandbox/scenarios/w02-install-supervised.yaml+++ b/beacon-sandbox/scenarios/w02-install-supervised.yaml@@ -25,12 +25,10 @@ install:expect:- action: prompt.submittedcontains: ["{{canary}}"]- optional: truewhy: >-- Recorded rather than gated, and for a narrower reason than it used to be. w01 and w03 have both- captured through installed hooks in system mode, so the hook command itself works on Windows.- What no run has yet observed is capture in *user* mode: the only attempt failed before the- session on an OTLP port conflict with a system collector the previous scenario left running.- That is fixed in the runner rather than here, but this stays optional until a run has actually- demonstrated it -- promoting it on the strength of the system-mode result would be assuming the- thing this scenario exists to check.+ Capture through a user-mode endpoint, which is a different path from the system-mode scenarios:+ a supervised collector rather than a service, and a log in the user's own profile rather than+ under %ProgramData%. It stayed optional until a run demonstrated it, because the system-mode+ result did not imply it -- the first attempt never reached the session, dying on an OTLP port+ conflict with a collector the previous scenario left behind. With that fixed the scenario+ captured 33 events, so the gap it was hedging against is closed.