Run local panel review, judging, and synthesis via Claude CLI.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.Jolymmiles/fusion-mcp" MCP server from askskill: Run: claude mcp add 'io-github-jolymmiles-fusion-mcp' -- npx -y fusion-mcp
Use the fusion workflow for this task: design a user permission system for a small-to-mid SaaS product. First generate 3 different approaches, then judge their pros and cons, and finally produce a synthesized recommendation with implementation steps.
Returns multiple candidate approaches, a judged comparison, and a synthesized recommended design with an implementation plan.
Run panel->judge->synthesis on this API documentation draft: first provide revision feedback from clarity, completeness, and actionability perspectives, then synthesize an improved document structure and sample content.
Produces multi-perspective review feedback and a clearer, more complete rewritten documentation plan.
Use fusion mode to review this implementation idea: I want to add a caching layer to a Node.js service. First propose several strategies, then judge them on performance, complexity, and maintainability, and finally recommend the best option.
Returns candidate implementation strategies, comparative judgments, and a final recommended technical approach.
Run local multi-model analysis, synthesis, and build-check loops for coding tasks.
Query multiple AI models in parallel and get structured judged analysis.
Run background multi-model deliberation to produce higher-quality synthesized AI answers.
Connect MCP clients to Claude Code CLI for read-only second opinions.
Provide code review, fixes, testing, and refactoring help in Claude Code.
Report issues and integration problems between Claude and MCP.