host-panel
Facilitate research-grounded panels in roundtable, Oxford, and Socratic formats. Use when exploring contested topics from multiple angles. NOT for Q&A, code review, or real human opinion simulation.
Facilitate research-grounded panels in roundtable, Oxford, and Socratic formats. Use when exploring contested topics from multiple angles. NOT for Q&A, code review, or real human opinion simulation.
Quick Start
Section titled “Quick Start”Install:
npx skills add github:wyattowalsh/agents --skill host-panel -y -g --agent claude-code --agent codex --agent gemini-cli --agent antigravity --agent github-copilot --agent opencodeUse: /host-panel "topic" [format] [num-experts]
Works with Claude Code, Gemini CLI, and other agentskills.io-compatible agents.
What It Does
Section titled “What It Does”host-panel stages a source-aware deliberation across simulated intellectual positions. It is designed to expose cruxes, evidence gaps, and trade-offs without pretending that synthetic panelists represent real human opinion, polling, prediction, or truth by consensus.
Research Basis
Section titled “Research Basis”The skill includes references/debate-research.md, a research-to-rule matrix mapping AI debate and role-play findings to panel mechanics. It draws on multi-agent debate, STORM-style perspective discovery, role-playing evaluation, sycophancy research, and adversarial-persuasion failure modes.
| Mechanic | Why it exists |
|---|---|
| Independent first positions | Reduces first-answer anchoring before cross-talk |
| Bounded critique | Avoids endless, circular debate rounds |
| Anti-conformity checks | Prevents majority pressure from masquerading as correctness |
| Source ledger | Separates verified claims from thin or unverified synthesis |
| Methodology cards | Makes panelists differ by evidence standard, not theatrical voice |
| Trajectory synthesis | Synthesizes by cruxes and evidence, not final-round vote |
Formats
Section titled “Formats”| Format | Purpose | Best for |
|---|---|---|
roundtable | Open multi-perspective mapping | Policy trade-offs, decision-critical topics, exploratory synthesis |
oxford | Testing a genuinely contestable proposition | Binary motions that pass the false-balance gate |
socratic | Probing definitions and assumptions | Conceptual disputes, ambiguous terms, hidden premises |
Defaults: roundtable, 4 experts. Expert range: 2-6.
Pipeline
Section titled “Pipeline”-
Parse — Preserve
/host-panel "topic" [format] [num-experts]; reject invalid formats, counts outside 2-6, and malformed topics. -
Diagnose — Classify the topic as settled-factual, open-value, controversial-factual, speculative, thin-evidence, decision-critical, or casual-experiential.
-
Ground — Build a source ledger when tools are available; otherwise label the panel as
unverified-training-knowledge. -
Map perspectives — Choose intellectual traditions from the terrain map and source ledger before creating personas.
-
Run the panel — Require independent first positions, bounded critique, moderator challenge, convergence checks, and format-specific phases.
-
Synthesize — Separate factual claims, interpretive disagreements, cruxes, evidence gaps, decision implications, and model limits.
Dispatch
Section titled “Dispatch”| Input | Mode |
|---|---|
Empty $ARGUMENTS | Show gallery, defaults, supported formats, and ask for a topic |
"topic" only | Parse topic, auto-select format, default to 4 experts |
"topic" roundtable|oxford|socratic | Use explicit format |
"topic" ... <2-6> | Use explicit expert count |
| Invalid format or count | Reject before research or personas |
| Loaded premise | Challenge the framing and offer a neutral motion |
| Settled factual issue | Reframe or decline false-balance debate |
| Q&A, code review, real opinion simulation | Redirect out of scope |
Output Contract
Section titled “Output Contract”Complete panels contain:
Topic GateResearch StatusSource LedgerTerrain MapPanelist RosterPanel PhasesFinal SynthesisFollow-up Options
Reference Files
Section titled “Reference Files”| File | Purpose |
|---|---|
references/topic-diagnostic.md | Topic classification, false-balance gate, and format selection |
references/research-integrity.md | Source ledger, citation integrity, confidence labels, and provenance |
references/debate-research.md | AI debate, role-play, sycophancy, and synthesis research mapped to rules |
references/archetypes.md | Methodology-card construction and anti-clustering guidance |
references/moderator-rules.md | Independent first positions, anti-conformity, turn-taking, and provocation |
references/formats.md | Roundtable, Oxford, and Socratic phase structures |
references/synthesis.md | Trajectory-aware final synthesis requirements |
Quality Gates
Section titled “Quality Gates”The current eval suite covers weak-topic reframing, false-balance handling, loaded premises, invalid arguments, thin evidence, citation traps, human-simulation traps, majority pressure, persuasive falsehoods, cultural/normative scope, and out-of-scope code-review redirects.