Coordinate multi-model councils to validate plans and consolidate judge verdicts.
npx skills add https://github.com/boshu2/agentops --skill council-boshu2
Please help me install this Agent Skill. Skill: council Source: https://github.com/boshu2/agentops/tree/main/skills/council Command: npx skills add https://github.com/boshu2/agentops --skill council-boshu2
This Skill coordinates a multi-model council to validate plans, brainstorm options, and perform research by spawning parallel judges and consolidating their verdicts.
Use /council validate recent to trigger a multi-model review and surface key findings.
High-intent search queries and answers about installing and using this skill.
Multi-model consensus validation works by spawning parallel AI judges with configurable perspectives to independently evaluate a plan, then consolidating their verdicts into a single structured JSON decision for downstream automation.
The best way to orchestrate multiple AI models for brainstorming and research is to use a council framework with configurable judge modes and optional explorer sub-agents, consolidating diverse perspectives into a unified verdict.
You can start a multi-model council to validate recent outputs by triggering a quick-start preset command, which immediately spawns parallel judges to review and surface key findings from the provided plans.
Yes, you need explicit frontmatter containing the name and description in SKILL.md to properly configure the multi-model council, alongside optional directories for scripts, references, and assets.
Yes, you can use predefined perspective configurations and quick-start presets to rapidly initiate common multi-model analyses, enabling diverse viewpoints for rigorous validation, brainstorming, and deep research tasks.
You should avoid using a multi-model council for task validation when your objective does not require diverse perspectives, rigorous validation, or deep analysis, or when downstream automation cannot process a structured JSON verdict schema.
BailipaCreate agent skills with standardized SKILL.md frontmatter and resource directories.
masharrattConsolidate CFN Loop agent outputs into structured Loop3Result and Loop2Result JSON.
hongbietcodeFacilitate multi-agent debate across Researcher, Critic, and Synthesizer roles.
danielgapRuns two parallel LLM reviewers to produce a fused verdict report.
