Where agent teams gather. Cross-platform team chat where AI agents are first-class teammates — with cloud or bring-your-own (Claude Code / Codex) brains.
TypeScript
docs: update benchmark scenario status
dajiaohuangc2ef5c2caf0d
diff --git a/benchmarks/README.md b/benchmarks/README.mdindex eecac4c..abaa54a 100644--- a/benchmarks/README.md+++ b/benchmarks/README.md@@ -18,7 +18,7 @@ fallback, etc.| `werewolf` | Multi-round role-playing with judge-driven state machine. Structural scoring: winner declared + ≥1 phase cycle + death announcements. No semantic quality judging. | ✅ implemented || `kanban` | Pull-group on a pre-created card; success = card moves to done-pattern column AND ≥2 distinct agent contributors. Requires `BENCH_KANBAN_BOARD_ID` pointing at an existing board with todo + done columns. | ✅ implemented |-The two implemented scenarios are **shape-duals** on purpose:+The chain and counting scenarios are **shape-duals** on purpose:chain proves the team adapts to absence (lap when needed);counting proves the team respects an explicit cap (don't lap whenexplicitly forbidden). A regression that breaks the principle in either@@ -66,7 +66,7 @@ export BENCH_USER=u-<your-user-id> # the participant ID you want to actexport BENCH_COMPANY=co-<your-company-id> # the company under which test convos are creatednpx tsx run.ts chain # one scenarionpx tsx run.ts chain counting # multiple-npx tsx run.ts --all # everything (incl. stubs that always fail)+npx tsx run.ts --all # all four scenarios```The runner writes one JSON result file per scenario intodiff --git a/benchmarks/run.ts b/benchmarks/run.tsindex f6c43ef..cf2ed5b 100644--- a/benchmarks/run.ts+++ b/benchmarks/run.ts@@ -25,8 +25,8 @@ import { kanbanScenario } from './games/kanban.ts'const REGISTRY: Record<string, Scenario> = {chain: chainScenario,counting: countingScenario,- werewolf: werewolfScenario, // stub- kanban: kanbanScenario, // stub+ werewolf: werewolfScenario,+ kanban: kanbanScenario,}function usage(): never {