agentleFS
bladecoderbladecoder/narrative-skills/skills/manuscript-review/SKILL.md

manuscript-review

Skill0 starsChanged 7 months agoMITView on GitHub

Review a partial or full manuscript across multiple chapters once enough draft material exists to judge cumulative structure, pacing, character arcs, thematic coherence, and chapter-to-chapter flow.

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. 0 are common, 13 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 13

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 14
  2. Very specific:

    Do not use this skill for isolated chapter feedback.

    line 22
  3. Distinctive:

    Ask questions only when the manuscript scope or review focus is genuinely unclear.

    line 36
  4. Very specific:

    all draft chapters in scope from 05_draft/

    line 43
  5. Distinctive:

    Read these when present and relevant:

    line 46
  6. Distinctive:

    If the project already distinguishes manuscript review snapshots by scope or date, follow that pattern instead of overwriting blindly.

    line 66
  7. Very specific:

    Read story.language before writing the manuscript review.

    line 72
  8. Distinctive:

    Keep file and folder names in English.

    line 73
  9. Distinctive:

    Write the review content in the selected story language.

    line 74
  10. Distinctive:

    Use correct orthography, accents, and normal Unicode for that language.

    line 75
  11. Very specific:

    If story.language is missing, gather it before writing.

    line 76
  12. Very specific:

    If the review file uses a different language than story.language, flag the mismatch and resolve it deliberately.

    line 77
  13. Distinctive:

    Read the manuscript material in scope in sequence, not as isolated excerpts.

    line 81
  14. Distinctive:

    After running the skill, provide:

    line 132
  15. Distinctive:

    Do not reduce the review to a pile of chapter summaries.

    line 151
  16. Distinctive:

    Do not rewrite the manuscript here.

    line 152
  17. Distinctive:

    Do not make structure recommendations that contradict the documented project intent without stating the conflict.

    line 153
  18. Distinctive:

    Do not ignore accepted chapter revisions when judging current state.

    line 154

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Manuscript Review

Overview

Use this skill to perform an editorial review at manuscript level rather than chapter level. It should evaluate the cumulative reading experience across several chapters, identify systemic problems, and produce a review artifact that helps decide whether to continue drafting, revise specific chapters, or restructure part of the book.

When To Use

Use this skill when:

  • several drafted chapters already exist;
  • the project needs a higher-level review than chapter-review;
  • pacing, escalation, arc progression, or thematic coherence must be assessed across chapters;
  • the user wants to know whether the manuscript is accumulating effectively;
  • local chapter quality is not enough to determine whether the book is working globally.

Do not use this skill for isolated chapter feedback. Use chapter-review for that.

Required Inputs

Gather or infer these inputs:

  • story.yaml metadata;
  • the chapter range or manuscript scope to review;
  • draft chapters in scope;
  • chapter reviews and revisions if they affect interpretation;
  • outline and chapter list;
  • core style and character files;
  • continuity and open-loop records when they affect cumulative judgment.

Ask questions only when the manuscript scope or review focus is genuinely unclear.

Files To Read

Always read:

  • story.yaml
  • all draft chapters in scope from 05_draft/
  • 02_story/chapter_list.md

Read these when present and relevant:

  • 00_meta/premise.md
  • 00_meta/goals.md
  • 00_meta/constraints.md
  • 01_style/*
  • 03_characters/*
  • 04_outline/*
  • 06_review/chapter_*.review.md
  • 06_review/chapter_*.revision.md
  • 02_story/open_loops.md
  • 02_story/synopsis.md
  • 07_continuity/*

Files To Write

Write or update:

  • 06_review/manuscript.review.md

If the project already distinguishes manuscript review snapshots by scope or date, follow that pattern instead of overwriting blindly.

Language Rules

This skill depends on story.yaml as the source of truth for project metadata.

  • Read story.language before writing the manuscript review.
  • Keep file and folder names in English.
  • Write the review content in the selected story language.
  • Use correct orthography, accents, and normal Unicode for that language.
  • If story.language is missing, gather it before writing.
  • If the review file uses a different language than story.language, flag the mismatch and resolve it deliberately.

Procedure

  1. Read the manuscript material in scope in sequence, not as isolated excerpts.
  2. Reconstruct the intended story route from the premise, chapter list, outline, and existing review artifacts.
  3. Evaluate the manuscript across cumulative dimensions such as:
    • hook and narrative promise;
    • escalation and pacing;
    • chapter-to-chapter propulsion;
    • coherence of character arcs;
    • thematic reinforcement;
    • exposition management;
    • clarity of stakes;
    • tonal consistency;
    • open-loop handling;
    • structural alignment between draft and outline.
  4. Distinguish between:
    • systemic problems affecting many chapters;
    • local problems best fixed chapter by chapter;
    • strengths that should be protected during revision.
  5. Prioritize the minimum revision strategy that materially improves the manuscript.
  6. Recommend whether the next move should be:
    • keep drafting;
    • rewrite specific chapters;
    • re-plan a section;
    • perform continuity or state sync before proceeding.

Output Requirements

The manuscript review should include:

  1. scope reviewed;
  2. overall assessment;
  3. what is working at manuscript level;
  4. major systemic problems;
  5. secondary or local issues;
  6. structural and pacing observations;
  7. character-arc observations;
  8. continuity or coherence risks at manuscript level;
  9. prioritized next actions.

The review should favor actionable synthesis over chapter-by-chapter repetition.

Quality Bar

The resulting review should allow another agent or editor to answer:

  • whether the manuscript is accumulating effectively;
  • which problems are structural rather than local;
  • what strengths should not be damaged by revision;
  • what the next highest-leverage intervention is.

Output

After running the skill, provide:

  • a short summary of the manuscript-level verdict;
  • the main systemic risks found;
  • whether the manuscript should continue drafting or pause for revision;
  • the recommended next step.

Definition Of Done

This skill is done when:

  • 06_review/manuscript.review.md exists and is useful;
  • systemic and local issues are clearly separated;
  • recommendations are concrete enough to drive the next workflow step;
  • the review respects story.language;
  • the manuscript's global state is clearer than before the review.

Restrictions

  • Do not reduce the review to a pile of chapter summaries.
  • Do not rewrite the manuscript here.
  • Do not make structure recommendations that contradict the documented project intent without stating the conflict.
  • Do not ignore accepted chapter revisions when judging current state.

Example Invocation

Use $manuscript-review to review the current manuscript across chapters 1 to 8, focusing on escalation, cumulative tension, and whether the protagonist's arc is developing cleanly.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.