Convene the multi-LLM round table — Claude, Antigravity, and Codex deliberate a question; the user chairs promotion. Triggers on: roundtable, round table, convene the table, ask the table, what do the other models think, deliberate.
Scanned 9/2/2026
Install to Claude Code
Skills Directorynpx -y skills add Smart-AI-Memory/attune-ai --skill roundtable --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Roundtable?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/smart-ai-memory-roundtable-attune-ai)More formats (shields.io, HTML) on the badges page.
---
name: roundtable
description: "Convene the multi-LLM round table — Claude, Antigravity, and Codex deliberate a question; the user chairs promotion. Triggers on: roundtable, round table, convene the table, ask the table, what do the other models think, deliberate."
argument-hint: "<question | read <thread> | promote <thread> | routine <name>>"
---
# Round Table (project shim)
This is the in-repo projection of the plugin skill so `/roundtable`
works in attune-ai sessions before the next plugin release ships
it. The canonical skill body is the plugin copy — Read
[plugin/skills/roundtable/SKILL.md](../../../plugin/skills/roundtable/SKILL.md)
now and follow it exactly. Do not duplicate its content here; if
this shim and the plugin copy ever disagree, the plugin copy wins.
The canonical workflow uses the shared command-workspace tools when
available, including atomic promotion batches of at most three candidates and
the one-candidate compatibility fallback.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
