AgoraX

Cross-disciplinary embracing AI era, learning Agent development. Thanks OpenAgents platform!

AgoraX targets review scenarios like PRD, architecture proposals, business memos, and incident post-mortems that 'need quick conclusions', solving problems of inconsistent standards, easily missed key points, and hard-to-trace conclusions in group chat reviews. Innovation: Introducing 'router-orchestrated serial multi-Agent collaboration': preparer first structures materials and raises key info gaps and clarifying questions, debater outputs pros/cons and risks/countermeasures based on same structure, reviewer summarizes with Go/No-Go/Revise ruling and action items, generating machine-readable decision_matrix scoring matrix.

Agent Planning: 4 Agents - router (host/orchestrator) listens to project chat events, maintains stage state (idle → waiting_preparer → waiting_debate → waiting_reviewer → done), names other Agents in sequence, caches intermediate artifacts. preparer (prep/structuring) extracts materials into structured outline, points out info gaps, gives clarifying questions (Top 3). debater (debater/risk opposition) outputs advantages/disadvantages, risks and mitigations, pending questions based on structured input. reviewer (ruling/summary) gives final conclusion (Go/No-Go/Revise), action items, produces machine-readable decision_matrix JSON for automation or tracking. OpenAgents collaboration: Serial Router Pattern - user posts materials in project chat → router @preparer → preparer replies → router @debater → debater replies → router @reviewer → reviewer outputs final result, flow ends.

① router (orchestrator): Listens to project chat events, recognizes user trigger info; maintains flow stage state, names other Agents in sequence, aggregates/caches intermediate results before posting to project chat. ② preparer (structuring): Extracts user's PRD/proposal/post-mortem into structured outline; outputs key info gaps (Top 3) and clarifying questions (Top 3), providing unified input for subsequent review. ③ debater (multi-perspective argumentation): Produces pros/cons, key risks and mitigations, pending open questions based on preparer's structured content, forming 'debatable' review materials. ④ reviewer (final ruling): Synthesizes preparer + debater output, gives Go/No-Go/Revise conclusion with action items; generates machine-readable decision_matrix JSON for tracking and comparing multiple review rounds. ⑤ OpenAgents collaboration (Serial Router Pattern): User posts materials → router @preparer → router @debater...