A sophisticated multi-model deliberation tool that gathers independent analysis from diverse AI providers to generate a synthesized consensus on complex problems.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install consilium
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install consilium using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Council?
The Council skill is designed for high-stakes decision-making where a single perspective is insufficient. By leveraging the Openclaw Skills framework, it assembles a virtual board of advisors consisting of 3-5 genuinely different AI models—such as Claude, GPT, and Gemini—rather than a single model role-playing different personas. Each model analyzes the user's query independently, ensuring that different training methodologies and reasoning patterns are utilized to catch potential blind spots.
This approach provides a level of rigor and depth that single-model interactions cannot match. Whether you are navigating technical architecture shifts or complex business strategies, the skill produces a comprehensive synthesis that highlights points of agreement, sharp disagreements, and valuable minority opinions, all within your existing agent environment.
Council Use Cases
- Evaluating major architectural changes, such as migrating from monoliths to microservices.
- Performing risk assessments for investment strategies or new business ventures.
- Resolving complex interpersonal or legal disputes, such as equity disagreements between founders.
- Stress-testing a hypothesis by assigning a contrarian lens to one of the AI panelists.
- Synthesizing deep research where cross-verification from multiple model providers is required.
How Council Works
- The user triggers the process using the /council command followed by a complex question or problem statement.
- The skill identifies available models and dispatches the query in parallel to 3-5 sub-agents, each assigned a specific role like Deep Thinker or Pragmatist.
- Each sub-agent generates an independent response based on its unique training data and the specific lens it was assigned.
- The skill monitors the progress and collects responses once the defined quorum is reached.
- A final synthesis is performed to identify the consensus, highlight diverging views, and provide a prioritized list of action items.
Council Setup
To use this within the Openclaw Skills ecosystem, ensure you have at least three different model providers configured in your environment.
- Initialize the panel configuration by running the command for the first time:
/council --models
- Review the generated
council-panel.jsonfile in your workspace root to confirm the assigned model slots. - Ensure your tool permissions allow for
sessions_spawnandsubagentsto enable parallel model execution.
Council Data Schema & Taxonomy
The skill organizes its configuration and deliberation data to ensure transparency and easy follow-up queries:
| Data Point | Storage Location | Description |
|---|---|---|
| Panel Config | council-panel.json |
Maps specific model IDs to slot roles like 'technical' or 'contrarian'. |
| Session Memory | Sub-agent history | Stores the full, un-synthesized response from each specific AI model. |
| Output Synthesis | Workspace Chat | The final Markdown report including the consensus status and action items. |
| Quorum Logic | Runtime State | Determines if enough models have responded to proceed with synthesis. |
Council Advanced Features
- Multi-round debates using the
--rounds 2flag to allow models to rebut each other's initial analysis. - Custom profiles (
thorough,balanced,fast) to optimize for either speed or depth of analysis. - Targeted follow-up capabilities allowing users to query specific panelists about their individual reasoning.
- Configurable quorum settings and timeouts to manage resource usage and response latency.
- Support for custom lenses and specific model assignments to tailor the expertise of the council.
SKILL.md
Loading
Related Openclaw Skills
Skills Audit (SkillLens)
A security-focused CLI tool designed to audit locally installed AI agent skills for exfiltration, execution, and policy risks.
Chromecast Control
A comprehensive CLI-based skill for discovering, casting media, and managing playback on Chromecast and Google Cast-enabled devices.
Tempo Workspace
A professional integration that connects Openclaw Skills to Tempo for real-time workspace awareness and automated insight extraction.
OpenClaw Config Backuper
An automated utility to backup OpenClaw configurations, local skills, and workspace data for disaster recovery and migration.
Icom IC-7610 Transceiver Control
A comprehensive automation interface for the Icom IC-7610 transceiver, enabling remote control, CW keying, and real-time telemetry monitoring.
Release Prep
A deep code audit and release preparation agent designed to automate the lifecycle of Python packages from testing to PyPI publishing.
Featured*
