{"skill":{"slug":"hyperthink","displayName":"Hyperthink","summary":"Triple-perspective deep research engine. Triggered by /hyperthink. Runs an interrogate flow to narrow scope, then executes a 6-stage fully automatic pipeline...","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":21,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1777354792460,"updatedAt":1777354823322},"latestVersion":{"version":"1.0.0","createdAt":1777354792460,"changelog":"Initial release of hyperthink: a fully automatic, triple-perspective council style deep research engine.\n\n- Provides an end-to-end research pipeline triggered by /hyperthink, with unattended multi-stage execution and no checkpoints after scope confirmation.\n- Orchestrates three agent personas (Optimist, Analyst, Critic) to generate sectioned deep-dives, fact-checks, and a unified synthesis.\n- Integrates with the interrogate skill for scope narrowing via adaptive Q&A before launch.\n- Outputs include per-section deep-dives, structured audits, a long-form synthesis, and an executive brief (.md and optional .docx).\n- Supports notifications via webhook or Telegram, with all outputs stored under /data/hyperthink/.\n- Requires Anthropic API and recommends python-docx for docx export. Estimated cost: ~$55–75/run. Typical run time: 8–18 hours.","license":"MIT-0"},"metadata":{"os":null,"systems":null},"owner":{"handle":"inxan3","userId":"s1777jy7gnm8xab1ce9cyb6vnd85ff93","displayName":"inxan3","image":"https://avatars.githubusercontent.com/u/208640470?v=4"},"moderation":null}