Compact the current conversation into a handoff document for another agent to pick up.
Use when user asks to read, summarize, or analyze a research paper (PDF or text). Triggers on keywords like "read paper", "summarize paper", "paper summary", "literature review", "analyze this paper". Produces either a Markdown summary or a styled HTML summary with hand-drawn SVG diagrams — when the user hasn't said which, ask first.
Adversarial code review using the opposite model. Spawns 1–3 reviewers on the opposing model (Claude spawns Codex, Codex spawns Claude) to challenge work from distinct critical lenses. Triggers: "adversarial review".
Update awesome-claude-code-config to the latest version. Checks remote for new releases, then re-runs the installer with the interactive selector. Use when user types /update-config or asks to update their Claude Code configuration.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.