Run a question, plan, decision, code review, or PR through a 5-advisor LLM Council with anonymized peer review and chairman synthesis. Each advisor uses a distinct REASONING METHOD (inversion, decomposition, analogy, naive questioning, dependency graphing) — not just a different persona. Based on Andrej Karpathy's LLM Council + DMAD research (ICLR 2025). MANDATORY TRIGGERS: 'council this', 'run the council', 'pressure-test this', 'stress-test this', 'war room this', 'debate this', 'convene the council'. DO NOT TRIGGER on simple factual questions, casual yes/no, lookups, or tasks with one obvious right answer. Reserve for decisions with genuine stakes and tradeoffs — cost is 11 sub-agent calls.
$npx -y skills add justsima/agentic-stack --skill llm-councilInstalls into the current project.
Run `npx skills use "https://github.com/justsima/agentic-stack" --skill "justsima/agentic-stack/llm-council"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/justsima/agentic-stack" that are relevant to the current task. Run `npx skills add "https://github.com/justsima/agentic-stack"` and select the relevant skills, then follow their instructions.