A headless integration for Openclaw Skills that delivers synthesized answers from a multi-model LLM council directly to your chat interface.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ask-council
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 ask-council 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 Ask Council?
Ask Council is a technical skill designed to provide high-quality, synthesized answers by leveraging the power of multiple AI models. Instead of relying on a single LLM, this skill initiates a full deliberation cycle via the LLM Council backend. It is a vital addition to the library of Openclaw Skills for users who require objective, cross-referenced intelligence without the need to switch to a web browser.
This skill is particularly effective for resolving complex queries where model bias or hallucination might be a concern. By utilizing a three-stage deliberation process, Ask Council ensures that the final output—the Chairman's synthesized answer—represents a consensus reached after models have responded and ranked each other.
Ask Council Use Cases
- Getting an objective consensus on technical decisions, such as architectural patterns or language choices.
- Accessing multi-model insights from mobile or constrained environments via Telegram or chat.
- Reducing model bias by forcing a debate between different LLM providers through Openclaw Skills commands.
- Quick fact-checking and logic verification using the synthesis stage of the LLM Council.
How Ask Council Works
- The user issues a query using the /council command followed by their question.
- The skill transmits the question to the LLM Council backend service.
- All available models generate individual responses during the first deliberation stage.
- Models evaluate and rank each other's responses to filter for the highest quality information.
- A final synthesis is performed by the Chairman model, which the skill then retrieves and displays to the user.
Ask Council Setup
To use this skill, the LLM Council backend must be active. You can initiate the backend installation with:
/install-llm-council
Once the backend is operational, Openclaw Skills can trigger the council deliberations. To run the query script manually via the terminal, use:
bash ~/.openclaw/skills/ask-council/ask-council.sh "your-question-here"
Ask Council Data Schema & Taxonomy
The Ask Council skill utilizes a straightforward file structure to manage its logic and metadata within the Openclaw Skills ecosystem:
| File | Role |
|---|---|
SKILL.md |
Primary documentation and user guide for the skill. |
ask-council.sh |
The core execution script that polls the API and extracts the chairman's synthesis. |
_meta.json |
Stores configuration data, including the version (1.0.3) and repository links. |
Ask Council Advanced Features
- Real-time polling with progress indicators to manage the 30-60 second deliberation window.
- Automated extraction of the Chairman's Answer from complex JSON deliberation payloads.
- Seamless hand-off to the web UI by providing direct URLs for deep-dive research into specific model disagreements.
- Stateless execution that ensures every request is treated as a fresh deliberation context for unbiased results.
SKILL.md
Loading
Related Openclaw Skills
Messari Crypto Intel
A comprehensive market intelligence skill for AI agents to access institutional-grade crypto data, real-time metrics, and AI-driven analysis via the Messari API.
ClawVille AI Agent Life Simulation
A persistent virtual world where AI agents work jobs, earn CLAW tokens, and level up through complex economic progression.
Nutrient Document Processing
A comprehensive document processing suite for OpenClaw that enables conversion, OCR, redaction, and digital signatures via natural language.
Jira
A comprehensive AI agent skill for managing Jira issues, sprints, and ticket workflows via CLI or MCP backends.
LLM Council Installer
A zero-config installer for the LLM Council app, enabling multi-model consensus and synthesis through a single command.
World Meeting Coordination Skill
A specialized scheduling engine that ranks meeting windows across timezones to find the most optimal overlap for distributed teams.
Featured*
