Roundtable AI

Multi-agent roundtable discussion — topic-driven multi-round debate with convergence detection and conclusion generation

Roundtable Discussion Skill

Overview

Enable multiple agents to participate in structured, multi-round discussions around a topic. Each participant brings their unique role perspective, and the system tracks convergence toward consensus.

Core value: Turn "one agent working alone" into "a team having a meeting."

When to Use

• Tech design review: product, frontend, backend, architect debate approach
• Competitive analysis: product, marketing, design compare alternatives
• Bug root cause analysis: backend, ops, test triangulate the issue
• Product requirements: product, design, dev align on scope
• Architecture decisions: architect, backend, frontend, devops choose stack

Prerequisites

Enable the roundtable toolset in the profile config:

toolsets:
  - roundtable

Or pass enabled_toolsets: ["roundtable"] when spawning an agent.

Tools

Tool	Purpose
roundtable_init	Create a discussion with topic + participants
roundtable_speak	Record a participant's speech
roundtable_read	Read discussion history
roundtable_status	Check status + convergence metrics
roundtable_summarize	Get structured data for conclusion doc
roundtable_end	Conclude or cancel a discussion
roundtable_list	List all discussions

Coordinator Flow

The coordinator (initiator) drives the discussion by orchestrating participants.

Step 1: Create the Discussion

roundtable_init(
    topic="Database selection: PostgreSQL vs MySQL vs TiDB",
    context="Our e-commerce system needs high-concurrency read/write, 1TB+ data",
    participants=[
        {"profile": "bingge", "role": "Product Director", "perspective": "Focus on UX", "display_name": "Bing"},
        {"profile": "mafei", "role": "Tech Lead", "perspective": "Focus on feasibility", "display_name": "Fei"},
        {"profile": "xiaosu", "role": "Designer", "perspective": "Focus on data display", "display_name": "Su"},
    ],
    max_rounds=3,
    speech_order="fixed"
)
→ returns {discussion_id, ...}

Step 2: Opening Statement (Round 0)

roundtable_speak(
    discussion_id="rt_xxxxxxxx",
    participant="coordinator",
    content="Today we're discussing database selection..."
)

Step 3: Multi-Round Discussion

For each round, read history then delegate to each participant:

# Read current state
history = roundtable_read(discussion_id="rt_xxxxxxxx")

# For each participant, delegate a sub-agent
delegate_task(
    goal="Participate in roundtable discussion as [ROLE]",
    context="You are participating in a roundtable discussion.\n\n"
            "Topic: {topic}\n"
            "Your role: {role}\n"
            "Your perspective: {perspective}\n\n"
            "Discussion history:\n{formatted_history}\n\n"
            "Please share your观点 using roundtable_speak. "
            "Keep it 200-500 words. Reference others' points if relevant."
)

Step 4: Check Convergence

After each round:

roundtable_status(discussion_id="rt_xxxxxxxx")
→ check convergence_score, consensus_points, disagreement_points

Step 5: Generate Conclusion

summary = roundtable_summarize(discussion_id="rt_xxxxxxxx")
→ Use the structured data to write a Markdown conclusion document
→ Save to the output_path specified during init

Step 6: End Discussion

roundtable_end(discussion_id="rt_xxxxxxxx")

Participant Prompt Template

When delegating to a participant sub-agent, use this template:

You are participating in a roundtable discussion.

## Discussion Info
- Topic: {topic}
- Context: {context}
- Current Round: Round {current_round} / {max_rounds}
- Your Role: {role} ({display_name})
- Your Perspective: {perspective}

## Discussion History
{formatted_history}

## Your Task
From your role's perspective, share your观点 on this topic.
- You may引用 or respond to other participants' statements
- Keep it concise and powerful, 200-500 words
- If you agree with a point, explicitly state your agreement
- If you disagree, explain why and propose alternatives

After speaking, call roundtable_speak to record your statement.

Convergence Detection

Each round is evaluated for convergence:

Metric	Formula	Meaning
Consensus	Points multiple participants agree on	Alignment
Disagreement	Points participants disagree on	Conflict
New Point	New topics raised this round	Scope expansion
Score	consensus / (consensus + disagreement)	Overall alignment

Termination conditions:

• Convergence score > 0.8 → high consensus, wrap up
• Max rounds reached → prevent infinite discussion
• Coordinator manually ends → emergency stop
• All participants vote to end → democratic close

Conclusion Document Format

# Roundtable Conclusion: [Topic]

## Summary
- Participants: Product(Bing), Design(Su), Dev(Fei)
- Rounds: 3
- Date: 2026-05-20

## Consensus Points
1. [Point 1]
2. [Point 2]

## Disagreement Points
1. [Point 1] - Various perspectives

## Action Items
1. [ ] [Action 1] - Owner: xxx
2. [ ] [Action 2] - Owner: xxx

## Detailed Transcript
### Round 1
- **Product(Bing)**: ...
- **Design(Su)**: ...
- **Dev(Fei)**: ...

### Round 2
...

Data Storage

• Database: ~/.hermes/roundtable.db (independent from kanban.db)
• Conclusion docs: Configurable via output_path, defaults to project docs dir
• ID format: rt_ + 8 hex chars (e.g., rt_a1b2c3d4)

Integration with Kanban

Discussions can be linked to kanban tasks:

# After conclusion, add as task comment
kanban_comment(task_id="t_xxx", body="Roundtable conclusion: {conclusion_path}")

Pitfalls

1. At least 2 participants required — A discussion needs multiple viewpoints
2. Participant must be registered — Only profiles listed in participants can speak
3. Round 0 is opening — Coordinator speaks first, then round 1 begins
4. Auto-conclude on max_rounds — Discussion ends automatically when max rounds exceeded
5. Independent database — roundtable.db is separate from kanban.db; don't mix paths
6. No LLM in summarize — roundtable_summarize returns raw data; the coordinator agent writes the conclusion