edit
SkillDev toolsRun line-level editing across chapters. Catches spelling, grammar, awkward phrasing, and word echoes.
Use edit in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add edit and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the edit skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
What this skill tells your AI
The instructions your AI receives, as published by howells/fiction in skills/edit/SKILL.md and read by Ahel’s review.
Run the editor agent to polish your manuscript at the line level.
What This Does
- Processes chapters looking for mechanical issues
- Flags spelling, grammar, awkward phrasing, word echoes
- Identifies manuscript-wide patterns (overused words, crutch phrases)
- Offers to apply mechanical fixes automatically
- Updates
progress.mdwith findings
Usage
/fiction:edit # Edit most recent chapter
/fiction:edit 5 # Edit chapter 5
/fiction:edit all # Edit all drafted chapters
/fiction:edit 3-7 # Edit chapters 3 through 7
If arguments provided: $ARGUMENTS
Parallel Processing (Important for Large Manuscripts)
When editing multiple chapters ("all" or a range), spawn editor agents in parallel for efficiency:
- Identify all chapters to process
- Launch one editor agent per chapter simultaneously using the Task tool
- Each agent processes its chapter independently
- After all complete, aggregate manuscript-wide patterns from all reports
- Update
progress.mdwith combined findings
Example parallel approach for /fiction:edit all with 20 chapters:
- Spawn 20 editor agents in a single message (one Task call per chapter)
- Agents run concurrently, each producing its own report
- Main conversation aggregates: common crutch words, repeated issues across chapters
- Total time ~ time for 1 chapter instead of 20x
What It Catches
- Style guide compliance — Enforces the project's declared style guide (Chicago Manual of Style for American, New Oxford Style Manual for British). Checks spelling conventions, punctuation rules, dash style, quotation marks, number rendering, capitalisation, and date formats. The editor reads
craft/tone.mdfor the project's declared guide, or detects it from context. - Spelling & typos — Including wrong-word errors (their/there) and variant-specific spelling (colour/color, travelled/traveled)
- Grammar & punctuation — Agreement, comma splices, dialogue punctuation per the active guide, serial comma usage, logical vs American quote placement
- Awkward phrasing — Confusing syntax, unclear references
- Word echoes — Repetition in close proximity
- Overused words — Filter words, weak verbs, crutch words
- Formatting — Dash style, ellipsis style, quote marks, scene breaks — all checked against the active style guide
- Audiobook readiness — Unattributed dialogue, ambiguous pronouns, similar-sounding names
Output
A report per chapter with:
- Issues by category with line numbers
- Suggested fixes
- Manuscript-wide patterns (when editing multiple chapters)
When to Use
- After completing a chapter draft
- Before sending to beta readers
- During revision passes
- After
/fiction:reviewaddresses story issues
Workflow
Recommended order for polishing:
/fiction:review— Fix story/craft issues first/fiction:edit— Then line-level polishcontinuityagent — Check cross-chapter consistency/fiction:critique— Final literary assessment (if complete)
Related Commands
/fiction:review— Story and craft feedback (run first)/fiction:critique— Full manuscript literary review/fiction:reconcile— Project structure audit
Signals
- GitHub stars
- 22
- Forks
- 6
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
edit-howells- Source
- github.com/howells/fiction