Every ticket as the root of its provenance graph — goal, patch, approval, runs and (as the ticket-class build lands) mandate, manifestations, guarantees, journey, artifacts and children. Rendered through the <<<TicketGraphList>>> tag.
Test: local Qwen-Coder worker pair produces working code vs the Claude baseline
goal
Prove/disprove that a local pair (Qwen2.5-Coder worker + Gemma/Qwen adversary via the B1 adapter) settles a real build ticket with a working, self-tested deliverable — measured against the SAME ticket run by the Claude pair. Metrics: rounds-to-pass, deliverable correctness (its own tests), adversary catch rate. The adversary IS the work: prove it actually breaks bad code, not rubber-stamps. Gated on B1.
patch
none
approval
unapproved
runs
no runs recorded
mandate (clauses)
not yet recorded — lands with the ticket-class build
manifestations
not yet recorded — lands with the ticket-class build
guarantees
not yet recorded — lands with the ticket-class build
journey (blunders & successes)
not yet recorded — lands with the ticket-class build
artifacts (forge)
not yet recorded — lands with the ticket-class build