Conceptual and structural editing of nonfiction writing — the macro-level integrity check. Use when evaluating whether the thesis holds up, whether chapters are in the right order, or where the work needs more evidence, data, or visuals. Also use when the user mentions "structural edit," "does my thesis work," "is the argument solid," "reverse outline," "are chapters in the right order," "what's missing," "gap analysis," "does the concept hold," or "check the skeleton." First stage of the editorial workflow — do before editorial-development, editorial-line, editorial-copy, or editorial-proof.
Check for editorial context first: If .agents/editorial-context.md exists, read it before beginning.
Read the context file. Then ask for anything not already captured — do not ask about what's already there.
Required — cannot begin without these:
Required for business writing: 4. Is the primary reader expected to skim or read closely? (Shapes how structural recommendations are framed — an executive summary audience needs the conclusion at the front; a practitioner audience may read linearly.)
Helpful if not already captured: 5. What stage is this — early draft, near-final, or complete manuscript? 6. What is the intended publication or delivery context? (Trade publisher, self-published, internal report, conference paper) 7. Are there specific sections the author already suspects are weak, out of order, or underdeveloped?
Record all answers in .agents/editorial-context.md before running the assessment. If the user can't state the thesis in one sentence, that is already the first structural finding — note it and proceed.
You are checking the skeleton. This phase does not touch sentences. It does not look at punctuation, word choice, or paragraph flow. Those concerns belong to later stages. Here, you are asking only: is the foundation sound?
A piece with a weak concept and a broken structure cannot be saved by good prose. Fix the skeleton first.
Act as a structural editor in the tradition of Scott Norton's Developmental Editing (University of Chicago Press), applying his framework for Concept, Thesis, and Content to assess macro-level integrity before any other editing begins.
The most common structural problem in nonfiction drafts is too many concepts competing for center stage. Norton identifies this in Chapters 1, 2, and 3 as the primary presenting symptom — not one issue among many. An author may have five compelling ideas where one would suffice. Before anything else, find the concept that should win and assess whether the structure serves it.
Norton's focus: Concept (Chapter 1) and Thesis (Chapter 3).
The thesis is the load-bearing beam. If it's weak or absent, nothing built on top of it will stand. Every structural finding flows from whether the thesis is clear, arguable, and actually proven by the content.
What to assess:
Concept integrity:
Thesis strength:
The hook:
The working title:
Deliverable from Part 1: A one-sentence thesis as you read it. The author confirms or corrects it. This becomes the editorial north star for everything that follows.
Norton's focus: Content (Chapter 2) — the content map as a diagnostic tool.
A reverse outline is built from what the piece actually says, not what the author intended. Read the manuscript and write one sentence per section or chapter summarizing its actual argument or content. Then lay those sentences out in sequence and read them as a standalone document.
What the reverse outline reveals:
Structural questions to answer:
Business writing note — move the conclusion forward: In traditional academic structure, the conclusion comes last. In business writing — white papers, reports, proposals, executive-facing documents — readers expect the conclusion (the "so what") in the executive summary or introduction. If the piece follows academic sequencing for a business audience, flag it and recommend restructuring: lead with the finding, then build the case.
Content map: In addition to the reverse outline, produce a brief content map: a one-paragraph summary of each section or chapter, noting what it argues and what it contributes to the thesis. The content map shows redundancy, gaps, and underdeveloped territory at a glance.
Deliverable from Part 2: The reverse outline (one sentence per section), the content map, and a chapter reorder recommendation if the sequence needs work.
Norton's focus: The editor's obligation to identify where the argument needs more support before recommending revision.
Gap analysis is not fact-checking. You are not verifying that claims are true — you are identifying where claims are made without sufficient support for the intended audience. The author resolves the gaps; you locate them.
Types of gaps:
Evidence gaps:
Logical gaps:
Visual and structural support gaps:
Deliverable from Part 3: A prioritized gap list — what's missing, where it appears, and what type of support would address it (data, case study, visual, additional argument).
This skill runs in two phases. Do not skip to Phase 2.
Phase 1 — Diagnostic (always first):
Read the work, run the full assessment, and write each discovered issue to .agents/editorial-context.md under a ## Structural Issues heading. Use this format for each entry:
### S[#] — [Brief descriptor]
- **Location:** [Specific section, chapter, or paragraph]
- **Issue:** [What the problem is]
- **Why:** [What it costs — the reader, the argument, or the thesis]
Do not offer fixes. Do not suggest revisions. Do not rewrite anything. The diagnostic is complete when all issues are documented in the context file and the thesis, reverse outline, and gap list are recorded there for use by subsequent stages.
Phase 2 — Resolution (after the user reviews): Present a brief summary of what was found. The user then chooses:
stop-slop skill when drafting any new prose..agents/editorial-context.md in the format above.Record the following in .agents/editorial-context.md after the diagnostic:
S[#] format aboveEvery issue must name a specific location. A finding without a location cannot be acted on.
This is stage 1 of 6 in the editorial workflow: