Research

Apply compaction, masking, and caching strategies

Git-aware undo by logical work unit (track, phase, or task)

Stanford STORM multi-perspective research method. Runs four sequential phases (multi-perspective scan, contradiction map, synthesis, peer review) in one thread to produce a PhD-level briefing on any topic. Use for \"storm research X\", \"multi-perspective research\", \"STORM method\", or any topic where a single-prompt answer is too shallow.

Long-running Jira investigation that self-continues past natural stops until a written hypothesis with evidence citations is produced. Pairs with the cooking Stop hook and hard-stops at 10 iterations.

Stanford STORM multi-perspective research method. Runs four sequential phases (multi-perspective scan, contradiction map, synthesis, peer review) in one thread to produce a PhD-level briefing on any topic. Use for \"storm research X\", \"multi-perspective research\", \"STORM method\", or any topic where a single-prompt answer is too shallow.

Multi-source deep research with confidence-rated synthesis. Use for \"research deeply\", \"deep dive on\", \"comprehensive research\", or /research-deep.

Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, product analytics warehouse) in parallel, then returns a cited read on decisions and tradeoffs. Use how for runtime behavior.

Convert vague requests into compact TCRO prompts with phase-specific constraints. Used by research, specify, plan, and work workflows.

Multi-phase research orchestration for thorough codebase, documentation, and external knowledge investigation. Invoked by /ai-eng/research command. Use when conducting deep analysis, exploring codebases, investigating patterns, or synthesizing findings from multiple sources.

Multi-phase research orchestration for thorough codebase, documentation, and external knowledge investigation. Invoked by /ai-eng/research command. Use when conducting deep analysis, exploring codebases, investigating patterns, or synthesizing findings from multiple sources.

Daily research agent for #research queue items — papers, concepts, authors. Runs the 5-step research loop with deeper source-finding and an Open Questions section.

Weekly research agent for #personal queue items — travel, products, life planning. Runs the 5-step research loop with a practical, actionable focus.

Daily research agent for #engineering queue items. Runs the 5-step research loop (Define → Triage → Query → Verify → Report) and writes a tiered, cited wiki page.

Premier research workflow for systematic investigation, synthesis, and verification. Use for literature reviews, competitive analysis, deep dives, market research, or any task requiring structured research with source attribution.

Multi-source deep research with confidence-rated synthesis. Use for \"research deeply\", \"deep dive on\", \"comprehensive research\", or /research-deep.

Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, product analytics warehouse) in parallel, then returns a cited read on decisions and tradeoffs. Use how for runtime behavior.

Multi-phase research orchestration for thorough codebase, documentation, and external knowledge investigation. Invoked by /ai-eng/research command. Use when conducting deep analysis, exploring codebases, investigating patterns, or synthesizing findings from multiple sources.

Multi-source deep research with confidence-rated synthesis. Use for \"research deeply\", \"deep dive on\", \"comprehensive research\", or /research-deep.

Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, product analytics warehouse) in parallel, then returns a cited read on decisions and tradeoffs. Use how for runtime behavior.

Multi-phase research orchestration for thorough codebase, documentation, and external knowledge investigation. Invoked by /ai-eng/research command. Use when conducting deep analysis, exploring codebases, investigating patterns, or synthesizing findings from multiple sources.

> Run 10 diagnostic checks on the knowledge base -- find orphans, drift, duplicates.

Synthesize accumulated observations into system evolution proposals. Fires when observation count reaches threshold. 5-phase process: triage, methodology updates, pattern detection, proposal generation, and human approval. Never auto-implements — always proposes changes for review. Triggers on: "rethink", "synthesize learnings", "evolve system", "improve process"

> Analyze the knowledge graph -- find connections, clusters, hubs, and synthesis opportunities.

Knowledge base from 'Science Research Writing for Non-Native Speakers of English' by Hilary Glasman-Deal (Imperial College London). Use when writing any section of a science research paper or thesis (Introduction, Methodology, Results, Discussion, Abstract), choosing verb tenses or passive/active voice, hedging causal claims, using a/the, signalling language, creating titles, or when asked to review/edit an academic paper's structure. 学术论文写作知识库：写论文各章节（引言/方法/结果/讨论/摘要）、选时态/被动语态、软化因果断言、用 a/the、起...