Agents execute. You create and decide.
runtime.trace
liveA Chinese mock-interview training workbench that starts from authorized voice transcription and closes the loop with question detection, evidence matching, and review.
A Chinese mock-interview training workbench that starts from authorized voice transcription and closes the loop with question detection, evidence matching, and review.
A public read-only analytics dashboard for multi-platform business media posts, covering publishing, engagement, and performance data.
A gaokao application simulator that uses province, track, score, or rank to suggest reach, match, and safety college-major options.
A unified API that lets AI agents reach users across WeChat, Telegram, and Lark, launch approvals, questions, and choices, and receive normalized response callbacks.
Use risk gates, scoring axes, and diagnostic questions to judge whether an AI startup idea deserves more work.
Audit prompts, skills, and agent definitions with an engineering checklist until issues converge.
Decompose complex build or design goals into goal docs, personas, staged plans, and validation loops.
Decide whether knowledge, rules, or retrospectives deserve a Codex skill and identify missing pieces.
Frame retrospectives or guardrails into the right skill domain, naming axis, boundaries, and acceptance bar.
Configure titles, descriptions, icons, and card images for stable Lark, Slack, and WeChat previews.
Get product updates, Skills, field notes, and co-build opportunities.
