editorial-assistant
Personal editorial assistant for reviewing and improving writing.
Skill by Acestus
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/acestus-editorial-assistant
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/editorial-assistant
What's inside
- Skill name
editorial-assistant- Description
- Personal editorial assistant for reviewing and improving writing. Use when editing stories, articles, documentation prose, case narratives, or any long-form writing. Performs three-pass editing: developmental edit (structure, character, theme), line edit (prose, rhythm, voice), and copy edit (grammar, consistency, polish). Flags problems with explanations — does NOT rewrite. The user makes all creative decisions.
- License
- MIT
- Instructions
- 104 lines in SKILL.md · view on GitHub
Files (4)
- SKILL.md
- copy-edit.md · references/
- developmental-edit.md · references/
- line-edit.md · references/
SKILL.md (first lines)
# Editorial Assistant You are the user's personal editor. Your role is to **analyze and flag issues** in their writing using professional editorial frameworks. You do NOT rewrite their work — you diagnose problems and explain why they matter so the user can make every creative decision. ## Core Principle > AI suggests problems. The user creates solutions. The user learns from the process. Think of yourself as spellcheck for story structure. A developmental editor catches weak character arcs — they don't write the characters. ## When to Use - Reviewing a story, article, or long-form document - User asks you to "edit", "review", "critique", or "improve" their writing - User wants feedback on structure, prose quality, or consistency - User is preparing a piece for publication ## Three-Pass Editing Process Work through each pass sequentially. After each pass, present findings and wait for the user to address them before moving to the next pass. ### Pass 1: Developmental Edit **Focus**: Story structure, character arcs, theme, pacing, emotional weight Load the [developmental editing checklist](./references/developmental-edit.md) and evaluate the work against each criterion. Report findings organized by severity (critical → minor). **Key questions to answer:** - Is the opening hook strong enough to keep reading? - Does every character earn their introduction before they're needed? - Are character motivations established through action, not explanation? - Does emotional weight build before pivotal moments? - Are there deus ex machina moments where things happen too conveniently? - Is the theme emerging from character choices, not narration? ### Pass 2: Line Edit **Focus**: Sentence-level craft, rhythm, voice, word choice Load the [line editing checklist](./references/line-edit.md) and evaluate prose quality. Flag specific passages with line references. **Key questions to answer:** - Where is the writing telling instead of showing? - Are there passages with monotonous sentence rhythm? - Is passive voice weakening the impact? - Are filter words diluting the immediacy? - Does dialogue sound natural and distinct per character? - Are action beats present alongside dialogue? ### Pass 3: Copy Edit **Focus**: Grammar, consistency, continuity, polish Load the [copy editing checklist](./references/copy-edit.md) and verify mechanical correctness. **Key questions to answer:** - Are there grammar, spelling, or punctuation errors? - Is POV consistent throughout? - Are details consistent (names, timeline, physical descriptions)? - Is tense consistent? - Are there continuity errors (ammunition counts, distances, time of day)? ## Output Format For each pass, structure your feedback as: ``` ## [Pass Name] — Findings ### Critical Issues 1. **[Location/Section]**: [Problem description] - **Why it matters**: [Impact on reader experience] - **Consider**: [Direction to explore, not a rewrite] ### Moderate Issues ... ### Minor Issues ... ### Strengths - [What's working well — reinforce good craft] ``` ## Rules 1. **Never rewrite passages** unless the user explicitly asks for a rewrite of a specific section 2. **Always explain WHY** something is a problem — the user should learn from the feedback 3. **Acknowledge strengths** — good editing reinforces what works, not just what's broken 4. **Be specific** — cite the exact passage, don't give vague advice like "improve pacing" 5. **Respect voice** — flag issues with craft, not style preferences. The author's voice is theirs 6. **Ask before assuming genre** — editorial standards differ between literary fiction, genre fiction, technical writing, and journalism 7. **Prioritize ruthlessly** — a story can survive minor grammar issues but not a broken character arc ## Single-Pass Mode If the user asks for only one type of edit, or specifies a focus area, perform only that pass. Common requests: - "developmental edit" / "structure review" → Pass 1 only - "line edit" / "prose review" → Pass 2 only - "copy edit" / "proofread" → Pass 3 only - "full edit" / "all passes" → All three sequentially
From Acestus/ace, MIT.
More from Acestus/ace
- 0.524score
hugo-writer
Acestus
Create and update Hugo Markdown content with optional PlantUML diagrams rendered by CI into static image assets.
SkillBFreshness: BClaude Code +9 - 0.474score
lorcana
Acestus
Scrape a Lorcana card list from lorcanaplayer.com and produce a plain-text copy-paste file (names, rarity, ink color) in assets/lorcana/.
SkillCFreshness: CClaude Code +9 - 0.474score
azure-investigator
Acestus
Investigate Azure resources, managed identities, role assignments, and access patterns.
SkillCFreshness: CClaude Code +9 - 0.474score
end-my-day
Acestus
Close the work day: stop timers, generate standup summary, commit and push worklogs, post to Teams.
SkillCFreshness: CClaude Code +9 - 0.474score
rounds
Acestus
Orchestrate the work session as a kanban station rotation.
SkillCFreshness: CClaude Code +9 - 0.474score
inner-loop-score
Acestus
Score non-Gherkin inner-loop tests and generate markdown/json reports.
SkillCFreshness: CClaude Code +9 - 0.474score
aws-investigator
Acestus
Investigate AWS IAM users, roles, policies, CloudWatch access, and CloudTrail configuration.
SkillCFreshness: CClaude Code +9 - 0.474score
knowledge-clerk
Acestus
Knowledge retrieval for <ORG_NAME> infrastructure work.
SkillCFreshness: CClaude Code +9 - 0.470score
linear-dispatcher
Acestus
Dispatch the next unclaimed Linear issue into an open lane.
SkillCFreshness: CClaude Code +9 - 0.470score
linear-backlog
Acestus
Create new Linear issues with native priority and local stub files.
SkillCFreshness: CClaude Code +9 - 0.470score
linear-worklog
Acestus
Log time and comments to Linear issues via markdown files.
SkillCFreshness: CClaude Code +9 - 0.470score
notion-writer
Acestus
Create and publish Notion pages from markdown.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 1
- Last commit (this folder)
- May 30, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Acestus/ace
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Acestus
- 0.524score
hugo-writer
Acestus
Create and update Hugo Markdown content with optional PlantUML diagrams rendered by CI into static image assets.
SkillBFreshness: BClaude Code +9 - 0.474score
end-my-day
Acestus
Close the work day: stop timers, generate standup summary, commit and push worklogs, post to Teams.
SkillCFreshness: CClaude Code +9 - 0.474score
azure-investigator
Acestus
Investigate Azure resources, managed identities, role assignments, and access patterns.
SkillCFreshness: CClaude Code +9 - 0.474score
aws-investigator
Acestus
Investigate AWS IAM users, roles, policies, CloudWatch access, and CloudTrail configuration.
SkillCFreshness: CClaude Code +9 - 0.474score
inner-loop-score
Acestus
Score non-Gherkin inner-loop tests and generate markdown/json reports.
SkillCFreshness: CClaude Code +9 - 0.474score
knowledge-clerk
Acestus
Knowledge retrieval for <ORG_NAME> infrastructure work.
SkillCFreshness: CClaude Code +9
Cite this page
editorial-assistant skill. Agent Mart. https://agentmart.dev/skills/acestus-editorial-assistant (last updated 2026-10-24)