brycewang-stanford/write

Write

Draft paper sections or apply humanizer pass by dispatching the Writer agent.

Input: $ARGUMENTS — section name or mode, optionally followed by file path.

---

Modes

/write [section] — Draft Paper Section

Draft a specific section: intro, strategy, results, conclusion, abstract, or full.

Agent: Writer Output: LaTeX section file in paper/sections/

Workflow:

1. Context Gathering

Before drafting, read all available context:

1. Read existing paper draft in paper/ (if it exists)
2. Read master_supporting_docs/ for notes, outlines, research specs
3. Read most recent quality_reports/research_spec_*.md or quality_reports/lit_review_*.md
4. Read .claude/references/domain-profile.md for field conventions
5. Check Bibliography_base.bib for available citations
6. Scan paper/tables/ and paper/figures/ for generated output
7. Read quality_reports/results_summary.md if it exists (from Coder)

2. Section Routing

Based on $ARGUMENTS:

• full: Draft all sections in sequence, pausing between major sections for user feedback
• intro: Draft introduction (most common request)
• strategy: Draft identification and estimation section
• results: Draft results from available output
• conclusion: Draft conclusion
• abstract: Draft abstract (must have other sections first)
• No argument: Ask user which section to draft

3. Dispatch Writer

Dispatch Writer with section standards, notation protocol, and anti-hedging rules. Humanizer pass runs automatically before finalizing. Save to paper/sections/[section].tex.

4. Quality Self-Check

Before presenting the draft:

• [ ] Every displayed equation is numbered (\label{eq:...})
• [ ] All \cite{} keys exist in Bibliography_base.bib
• [ ] Introduction contribution paragraph names specific papers
• [ ] Effect sizes stated with units
• [ ] No banned hedging phrases
• [ ] Notation consistent throughout
• [ ] All tables/figures referenced actually exist in paper/tables/ or paper/figures/

5. Present to User

Present each section for feedback. Flag items that need attention:

• TBD items: Where empirical results are needed but not yet available
• VERIFY items: Citations that need user confirmation
• PLACEHOLDER items: Effect sizes awaiting final estimates

/write humanize [file] — Humanizer Pass Only

Strip AI writing patterns from existing text without rewriting content.

Agent: Writer (humanizer mode) Output: Edited file with AI patterns removed

Strips 24 patterns across 4 categories:

• Structural: forced narrative arcs, artificial progression
• Lexical: "delve,leverage,nuanced,robust"
• Rhetorical: rule-of-three, negative parallelisms, em dash overuse
• Formatting: excessive bullet points, promotional language

---

Section Standards

| Section | Length | Key Requirements | |---------|--------|-----------------| | Introduction | 1000-1500 words | Hook → question → method → finding → contribution → roadmap | | Strategy | 800-1200 words | Formal assumption, numbered equation, threats addressed | | Results | 800-1500 words | Main spec, effect sizes in economic terms, heterogeneity | | Conclusion | 500-700 words | Restate with effect size, policy, limitations, future | | Abstract | 100-150 words | Question, method, finding with magnitude, implication |

---

LaTeX Conventions

• \citet{} for textual citations ("Smith (2024) shows...")
• \citep{} for parenthetical citations ("...is well documented (Smith, 2024)")
• booktabs rules (\toprule, \midrule, \bottomrule) — never \hline
• Notation protocol: Y_{it}, D_{it}, \gamma_i, \delta_t, \varepsilon_{it}

---

Principles

• This is the user's paper, not Claude's. Match their voice and style.
• Never fabricate results. Use TBD placeholders.
• Citations must be verifiable. Only cite confirmed papers.
• Humanizer is automatic. Every draft gets de-AI-ified.