← Back to Skills Marketplace
qui-council
by
QuincyGunter
· GitHub ↗
· v1.0.0
· MIT-0
85
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install qui-council
Description
Send an idea to the Council of the Wise for multi-perspective feedback. Spawns sub-agents to analyze from multiple expert perspectives. Auto-discovers agent...
Usage Guidance
This skill appears to do what it says: it uses persona .md files in its agents/ folder to spawn sub-agents and synthesize multi-perspective feedback. It's generally safe to install, but be careful about adding or accepting third-party agent .md files into the agents/ folder — those files can change the behavior of spawned sub-agents (they could, for example, prompt for secrets or ask to post data elsewhere). Before sending highly sensitive content to the council, redact secrets or restrict the input. If you plan to add custom agents, review their content for instructions that request credentials, external endpoints, or file reads. If you need higher assurance, request a code-level review of any added agent files or require a signed/verified source for custom personas.
Capability Analysis
Type: OpenClaw Skill
Name: qui-council
Version: 1.0.0
The 'Council of the Wise' skill is a multi-agent orchestration tool designed to provide multi-perspective feedback on user ideas. It functions by auto-discovering persona files in a local 'agents/' directory and spawning a sub-agent to synthesize their outputs. Analysis of the SKILL.md instructions and the individual agent markdown files (e.g., Analyst.md, Architect.md) reveals no evidence of malicious intent, data exfiltration, or harmful prompt injection. The skill operates entirely within the expected bounds of the OpenClaw/SkillBoss platform's sub-agent capabilities.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description (multi-perspective council) match the actual behavior: discover persona .md files, spawn sub-agents, synthesize responses. The skill does not request unrelated binaries, env vars, or config paths.
Instruction Scope
SKILL.md confines runtime actions to: reading the skill's agents/ folder, spawning sub-agents with a 5-minute timeout, and returning synthesized feedback. That is proportionate to the stated purpose. Note: because the skill auto-discovers any .md in agents/, a malicious or poorly written persona file could instruct a sub-agent to ask for secrets, call external endpoints, or attempt exfiltration — this is a property of auto-discovery, not evidence of current malicious behavior. The bundled agent files included in the skill appear purpose-aligned and contain no network endpoints or credential requests.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest-risk install profile. Nothing is downloaded or written by an install step.
Credentials
Requires no environment variables, no primary credential, and no config paths. The declared environment footprint is minimal and appropriate for the skill's functionality.
Persistence & Privilege
always:false (default) and model invocation allowed (default). The skill does not request permanent presence or modify other skills' configs. Autonomous invocation is normal for skills; combined with the low privilege footprint this is acceptable.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install qui-council - After installation, invoke the skill by name or use
/qui-council - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
**Council skill is now auto-discoverable and easier to use with flexible expert input.**
- Automatically detects and includes council members based on agent files in the `agents/` folder—no config needed.
- Output revamped with a clear structure: moderator synthesis first, followed by detailed expert perspectives, prioritized action items, and a confidence signal.
- Easily add new expert roles by dropping `.md` files with YAML frontmatter into the `agents/` folder.
- Five-minute timeout ensures thorough yet timely council feedback, providing partial results on timeout.
- Ideal for in-depth review of business plans, projects, or major decisions—not for quick answers or simple tasks.
Metadata
Frequently Asked Questions
What is qui-council?
Send an idea to the Council of the Wise for multi-perspective feedback. Spawns sub-agents to analyze from multiple expert perspectives. Auto-discovers agent... It is an AI Agent Skill for Claude Code / OpenClaw, with 85 downloads so far.
How do I install qui-council?
Run "/install qui-council" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is qui-council free?
Yes, qui-council is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does qui-council support?
qui-council is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created qui-council?
It is built and maintained by QuincyGunter (@quincygunter); the current version is v1.0.0.
More Skills