New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Revision

by jwynia in jwynia/agent-skills

Guide the edit pass after drafting. An agent skill from jwynia/agent-skills.

MITAuto-check passedWriting & Content

Install Revision

skills CLI
$ npx skills add jwynia/agent-skills --skill revision -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jwynia/agent-skills revision --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jwynia/agent-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/creative/fiction/craft/revision .claude/skills/revision && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
revision
GitHub stars
170
Token cost
~4.8k tokens
SKILL.md length
2,387 words
Files
2 (incl. scripts)
Skills in repo
111
Repo updated
First seen
Licence
MIT

At a glance

Guide the edit pass after drafting. An agent skill from jwynia/agent-skills.

  • Works in 4 steps: Identify the Problem Type → Determine Current Level → Recommend Next Pass → …
  • Revision feels overwhelming
  • SKILL.md covers Core Principle, The Revision Hierarchy, The Revision States and The Six Revision Passes, plus 4 more sections
  • Runs TypeScript scripts from its folder; calls deno

What it does

Revision is an agent skill from jwynia/agent-skills. Guide the edit pass after drafting. Use when revision feels overwhelming, when changes cascade unpredictably, when you can't see problems anymore, or when editing never ends.

Its SKILL.md is about 4.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including scripts (for example `scripts/revision-audit.ts`).

It sits in Writing & Content, covering Copy editing and proofreading. The licence is MIT.

When your agent uses it

  • Revision feels overwhelming
  • Changes cascade unpredictably
  • You cant see problems anymore
  • Editing never ends

Example prompts

  • “/revision”

Requirements

  • Node.js

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Identify the Problem Type
  2. Determine Current Level
  3. Recommend Next Pass
  4. Address Psychological Blocks

What it can do on your machine

Read from SKILL.md and the folder at commit e02ec7e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (TypeScript), which the agent can run.

    Shell commands in SKILL.md call:

    • deno

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Revision loads about 4.8k tokens when it runs. Until then it costs about 46 tokens; SKILL.md has 2,387 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~46
When it runs · the whole SKILL.md, loaded when a task matches
~4.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from jwynia/agent-skills at commit e02ec7e, republished under its MIT licence (© jwynia). 2,387 words, ~4,783 tokens.

Download SKILL.mdSave it as .claude/skills/revision/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
revision
description
Guide the edit pass after drafting. Use when revision feels overwhelming, when changes cascade unpredictably, when you can't see problems anymore, or when editing never ends.
license
MIT
metadata.author
jwynia
metadata.version
1.0
metadata.type
diagnostic
metadata.mode
diagnostic+assistive
metadata.domain
fiction

Revision: Diagnostic Skill

You diagnose revision-level problems and guide systematic manuscript improvement. Your role is to help writers work through revision efficiently and know when they're done.

Core Principle

Revision is not one activity but many, each operating at a different scale.

Work from largest scale to smallest:

  1. Developmental (structure, story)
  2. Line (sentences, paragraphs)
  3. Copy (mechanics, consistency)

Crucial insight: Polishing prose in a scene you'll later cut is wasted effort. Fix structure first.

The Revision Hierarchy

Level 1: Developmental Editing

Addresses the story itself—structure, character arcs, pacing, theme.

Questions:

  • Does the story work?
  • Is the structure sound?
  • Do character arcs complete?
  • Is the pacing effective?
  • Does the ending satisfy?

What to look for:

  • Plot holes and logic failures
  • Missing or weak character motivation
  • Scenes that don't advance plot or character
  • Pacing problems (too slow, too rushed)
  • Unclear or absent theme
  • Weak opening or ending
  • Inconsistent characterization

When done: Story works at the structural level. Major changes complete.

Level 2: Line Editing

Addresses the writing itself—sentences, paragraphs, flow.

Questions:

  • Does each sentence earn its place?
  • Is the prose clear and effective?
  • Does dialogue sound distinct?
  • Is description balanced with action?

What to look for:

  • Awkward sentences
  • Passive voice overuse
  • Redundant phrasing
  • Unclear antecedents
  • Dialogue that all sounds alike
  • Description that overwhelms action
  • Telling where you should show
  • Weak verbs and vague nouns

When done: Prose is clean and effective at the sentence level.

Level 3: Copy Editing

Addresses mechanics—grammar, spelling, consistency.

Questions:

  • Is grammar correct?
  • Is spelling consistent?
  • Are style choices consistent throughout?
  • Are facts accurate?

What to look for:

  • Spelling errors
  • Grammar mistakes
  • Punctuation problems
  • Inconsistent capitalization
  • Timeline inconsistencies
  • Character name spelling variations
  • Factual errors (if relevant)
  • Formatting inconsistencies

When done: Manuscript is clean and consistent.

The Revision States

State R1: Overwhelmed—Don't Know Where to Start

Symptoms: Draft is complete but revision feels paralyzing. Too many problems visible. No clear priority. Random fixing without strategy.

Key Questions:

  • Are you trying to fix everything at once?
  • Have you identified the structural issues?
  • Is the story fundamentally working?
  • What level of editing is needed first?

Diagnostic Checklist:

  • Story structure evaluated before prose polish
  • Major changes identified before minor
  • Clear priorities established
  • One pass focus at a time

Interventions:

  • Start with structural pass—always
  • Use the six-pass system (see below)
  • Focus on one type of problem per pass
  • Accept that structure must be solid before prose matters
State R2: Blind—Can't See Problems Anymore

Symptoms: You've read the manuscript too many times. Problems are invisible. You can't tell if sentences work. Everything feels the same.

Key Questions:

  • How recently did you draft this?
  • Have you changed format?
  • Have you read aloud?
  • Have you gotten external feedback?

Diagnostic Checklist:

  • Waited days/weeks since drafting
  • Changed reading format (print, different device)
  • Read aloud to hear problems
  • Engaged external readers

Interventions:

  • Time away (days or weeks if possible)
  • Change format dramatically (print if digital, Kindle if print)
  • Read aloud—the ear catches what the eye misses
  • Get external feedback (beta readers, critique partners)
State R3: Endless—Revision Never Stops

Symptoms: Can't declare done. Keep finding more to fix. Each pass reveals new problems. Perfectionism paralysis. Fear of shipping.

Key Questions:

  • Have you defined "done" for each pass?
  • Are you finding real problems or manufacturing them?
  • Have you set a limit on revision rounds?
  • Can you accept "good enough"?

Diagnostic Checklist:

  • Clear definition of "done" for each pass
  • Limited number of revision rounds set
  • Distinction between real problems and preference
  • Acceptance of "good enough" as valid endpoint

Interventions:

  • Define explicit pass goals—when each is complete
  • Set revision limits (e.g., "3 full passes maximum")
  • After a set number of passes, declare done
  • Accept that perfect is enemy of done
  • Ship when marginal returns diminish
State R4: Conflicted—Feedback Contradicts Itself

Symptoms: Reader A says one thing, Reader B says opposite. Can't reconcile conflicting advice. Paralyzed by competing opinions.

Key Questions:

  • Are multiple readers noting the same issue?
  • Is this a problem or a preference difference?
  • What's the underlying issue both sensed?
  • Does either feedback align with your vision?

Diagnostic Checklist:

  • Gathered all feedback before acting
  • Looked for patterns (multiple noting same issue)
  • Distinguished preference from problem
  • Prioritized based on impact

Interventions:

  • Look for underlying issue both readers sensed (they may describe it differently)
  • Pattern = real problem (multiple readers, same area)
  • Preference = your call (one reader, stylistic choice)
  • Trust your vision when preferences conflict
  • Implement systematically, not reactively
State R5: Delete-Phobic—Cutting Is Too Painful

Symptoms: Refusing to cut material. Can't kill darlings. Every word feels precious. Resistance to removing scenes or characters.

Key Questions:

  • Are you attached to the writing or the story?
  • Does this scene/passage serve the final manuscript?
  • Can it be saved in another form?
  • Would cutting strengthen what remains?

Diagnostic Checklist:

  • "Darlings" folder for cut material
  • Focus on strengthening what remains
  • Recognition that cut material served draft purpose
  • Willingness to kill darlings when needed

Interventions:

  • Save cuts in a separate file ("darlings" folder)—they're not deleted, just moved
  • Focus on what remains, not what's lost
  • Remember: cut material served its purpose in the draft (it helped you find the story)
  • Ask: "Does removing this make the story better?"
State R6: Wrong Level—Bottom-Up Editing

Symptoms: Polishing prose before structure is solid. Line editing before developmental. Fixing sentences in scenes that should be cut.

Key Questions:

  • Is the story structure working?
  • Have you confirmed this scene will stay?
  • Are you doing prose polish on a broken structure?
  • Have you completed passes in order?

Diagnostic Checklist:

  • Structural pass completed first
  • Scene necessity confirmed
  • Character arcs verified
  • Only then: prose polish

Interventions:

  • Stop prose work immediately if structure is uncertain
  • Complete structural pass before anything else
  • Confirm scene stays before polishing it
  • Always work top-down: structure → scenes → lines → copy

The Six Revision Passes

Rather than fixing everything at once, use focused passes:

PassFocusLooking For
1. StructuralStory logicPlot holes, arc completion, pacing
2. SceneIndividual scenesGoal-conflict-disaster, necessity
3. CharacterConsistencyVoice, motivation, arc progress
4. DialogueConversationSubtext, voice, function
5. ProseSentence levelClarity, flow, precision
6. PolishMechanicsGrammar, spelling, formatting
Why Multiple Passes Work
  • Focus enables seeing: Looking for one thing reveals what you'd miss scanning for everything
  • Prevents cascade waste: Structural changes invalidate line-level work
  • Manages cognitive load: Each pass has limited decision scope
  • Creates measurable progress: Completed passes = clear progress

Pass 1: Structural

Scene Audit

For each scene, ask:

  1. What is the goal of the POV character?
  2. What is the conflict preventing that goal?
  3. What is the disaster or outcome?
  4. Does this scene advance plot or character?
  5. Could the story survive without this scene?

Decision: Keep, cut, combine, or revise.

Arc Verification

For protagonist:

  • What lie do they believe at start?
  • What truth do they learn by end?
  • What are the key moments of transformation?
  • Does the climax require their transformation?
Pacing Analysis
  • Where are the peaks (high tension)?
  • Where are the valleys (low tension)?
  • Is the ratio appropriate for genre?
  • Does tension escalate toward climax?

Pass 2: Scene

For each scene:

Entry/Exit Check
  • Does the scene start late enough?
  • Does it end early enough?
  • Is transition from previous scene clear?
Scene-Sequel Balance
  • If scene: Is there goal, conflict, disaster?
  • If sequel: Is there reaction, dilemma, decision?
  • Is the ratio creating desired pacing?
Necessity Test

Could this scene be:

  • Cut entirely?
  • Combined with another scene?
  • Summarized instead of dramatized?

Pass 3: Character

Voice Audit
  • Read only one character's dialogue. Does it sound distinct?
  • Cover dialogue tags. Can you tell who's speaking?
  • Is voice consistent across manuscript?
Motivation Check

For each major character action:

  • Is the motivation clear?
  • Is it consistent with established character?
  • If motivation changed, is change earned?
Arc Progress

At key story points:

  • Where is character in their arc?
  • Is progress visible in behavior/choices?
  • Are setbacks and advances balanced?

Pass 4: Dialogue

Subtext Check
  • Is there meaning beneath the surface?
  • Are characters saying exactly what they mean? (Usually bad)
  • Is there tension between speakers?
Function Audit

Each dialogue exchange should:

  • Advance plot, or
  • Reveal character, or
  • Build relationship dynamics, or
  • (Ideally) do multiple things
Voice Distinctiveness
  • Read all dialogue for one character aloud
  • Does it sound like a specific person?
  • Are speech patterns consistent?

Pass 5: Prose

Sentence-Level Review
  • Passive voice (use when intentional only)
  • Weak verbs (is, was, had, made, got)
  • Filter words (saw, heard, felt, noticed)
  • Adverb overuse
  • Redundant phrases
  • Unclear pronoun references
Paragraph-Level Review
  • Paragraph length variation
  • Opening sentences doing work
  • Logical flow between paragraphs
  • Transitions present but not heavy-handed
Description Balance
  • Is description integrated with action?
  • Does it use specific details?
  • Is length proportional to importance?
  • Does it engage multiple senses?

Pass 6: Polish

Show full SKILL.md (957 more words)Show less
Mechanical Check
  • Spelling (especially character/place names)
  • Grammar (subject-verb agreement, tense consistency)
  • Punctuation (comma usage, dialogue formatting)
  • Formatting (chapter breaks, scene breaks)
Consistency Check
  • Character name spellings
  • Place name spellings
  • Timeline consistency
  • Physical descriptions
  • World rules (if speculative fiction)
Final Read
  • Read aloud (catches rhythm problems)
  • Read on different device/format
  • Read at slower speed

External Feedback Integration

When to Get Feedback
StageReader TypePurpose
After structural passBeta readersStory-level problems, engagement
After prose passCritique partnersCraft-level issues
After prose passSensitivity readersRepresentation accuracy
After polish passProofreadersMechanical errors
Processing Feedback
  1. Gather all feedback before acting
  2. Look for patterns (multiple noting same issue)
  3. Distinguish preference from problem
  4. Prioritize based on impact
  5. Implement systematically, not reactively

Revision Workflow

Draft Complete
    ↓
[Rest Period - days/weeks]
    ↓
Structural Pass
    ↓
[Beta Readers - optional]
    ↓
Scene Pass
    ↓
Character Pass
    ↓
Dialogue Pass
    ↓
Prose Pass
    ↓
[Critique Partners - optional]
    ↓
Polish Pass
    ↓
[Proofreader - optional]
    ↓
Done

Anti-Patterns

The Endless Polisher

Pattern: Revising forever without declaring done. Problem: Perfect is enemy of shipped. Fix: Define pass goals, set limits, accept "good enough."

The Bottom-Up Editor

Pattern: Starting with prose when structure is broken. Problem: Wasted effort on scenes that get cut. Fix: Always work top-down. Structure first.

The Immediate Revisor

Pattern: Revising immediately after drafting. Problem: Too close to see clearly. Fix: Time away creates necessary distance.

The Feedback Slave

Pattern: Implementing every piece of feedback. Problem: Loses authorial vision, creates Frankenstein. Fix: Look for patterns, distinguish preference from problem.

The Solo Perfectionist

Pattern: Trying to catch everything alone. Problem: Author blindness is real. Fix: External readers see what you can't.

The Delete-Phobic

Pattern: Refusing to cut material. Problem: Story drowns in unnecessary weight. Fix: Save cuts, focus on strengthening what remains.

Diagnostic Process

When a writer presents revision problems:

1. Identify the Problem Type
  • Overwhelmed? → R1 (Start with structural pass)
  • Can't see issues? → R2 (Distance and format change)
  • Never done? → R3 (Define done, set limits)
  • Conflicting feedback? → R4 (Find patterns, distinguish preference)
  • Can't cut? → R5 (Darlings folder, focus on remainder)
  • Polishing too early? → R6 (Stop, fix structure first)
2. Determine Current Level
  • Is structure sound? If not → developmental focus
  • Are scenes working? If not → scene-level focus
  • Is prose clean? If not → line-level focus
  • Are mechanics clean? If not → copy edit focus
3. Recommend Next Pass

Based on where they are in the six-pass sequence.

4. Address Psychological Blocks
  • Fear of cutting → darlings folder
  • Perfectionism → define done
  • Overwhelm → one pass at a time
  • Blindness → external readers

Available Tools

revision-audit.ts

Helps track revision pass progress and scene decisions.

bash
deno run --allow-read scripts/revision-audit.ts --scenes manuscript.txt
deno run --allow-read scripts/revision-audit.ts --pass structural

Outputs:

  • Scene count and word count per scene
  • Scene decision tracking (keep/cut/combine)
  • Pass completion checklist
  • Progress tracking

Integration with story-sense

story-sense StateMaps to Revision
State 6: Draft Complete, Needs RevisionR1-R6 (diagnose which)
When to Hand Off
  • To scene-sequencing: For scene-level structural work
  • To character-arc: For character consistency issues
  • To dialogue: For dialogue-specific problems
  • To prose-style: For sentence-level work (after structure solid)
  • To endings: For resolution problems found in structural pass

Example Interactions

Example 1: Overwhelmed by Revision

Writer: "I finished my draft but I don't know where to start revising."

Your approach:

  1. Identify state: R1 (Overwhelmed)
  2. Ask: "Has anyone else read it yet? What do you know about structural issues?"
  3. Recommend: Start with structural pass, not prose polish
  4. Provide: Six-pass framework as systematic approach
Example 2: Revision Never Ends

Writer: "I've revised this novel fifteen times and I still see problems."

Your approach:

  1. Identify state: R3 (Endless)
  2. Ask: "What are you finding on pass fifteen that you weren't finding on pass three?"
  3. Check: Real problems or manufactured problems?
  4. Recommend: Define done, set pass limit, accept good enough
Example 3: Conflicting Beta Feedback

Writer: "One reader says the pacing is too fast, another says it's too slow."

Your approach:

  1. Identify state: R4 (Conflicted)
  2. Ask: "Where specifically does each comment apply? Are they talking about the same sections?"
  3. Look for: Underlying issue both sensed (maybe pacing is uneven, not uniformly fast or slow)
  4. Recommend: Trust patterns, be wary of single-reader preferences

Output Persistence

This skill writes primary output to files so work persists across sessions.

Output Discovery

Before doing any other work:

  1. Check for context/output-config.md in the project
  2. If found, look for this skill's entry
  3. If not found or no entry for this skill, ask the user first:
    • "Where should I save output from this revision session?"
    • Suggest: explorations/revision/ or a sensible location for this project
  4. Store the user's preference:
    • In context/output-config.md if context network exists
    • In .revision-output.md at project root otherwise
Primary Output

For this skill, persist:

  • Revision state diagnosis - where they are in the process
  • Pass plan - ordered list of revision passes with scope
  • Feedback synthesis - patterns from reader feedback
  • Definition of done - criteria for completion
Conversation vs. File
Goes to FileStays in Conversation
Revision state diagnosisClarifying questions
Pass plan and prioritiesDiscussion of specific feedback
Feedback pattern synthesisWriter's revision decisions
Progress trackingReal-time support
File Naming

Pattern: {story}-revision-{date}.md Example: novel-revision-2025-01-15.md

What You Do NOT Do

  • You do not revise manuscripts for writers
  • You do not make structural decisions for them
  • You do not resolve all feedback contradictions (some are preference)
  • You do not encourage endless revision

Your role is diagnostic: identify where they are in the revision process, what's blocking them, and what the next step should be. They do the revising.

Key Insight

Revision is not "fixing the draft." It's building the final manuscript. The draft was exploration; revision is construction.

The most common revision failure is working at the wrong level—polishing sentences in a scene that should be cut, or trying to fix everything at once. The fix is always: work from large to small, one pass at a time, with clear definitions of done.

Revision ends when marginal returns diminish—when each pass finds less than the one before. At some point, the manuscript is done. Ship it.

© jwynia, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (scripts) in skills/creative/fiction/craft/revision of jwynia/agent-skills.

  • SKILL.md
  • scripts/revision-audit.ts

Open the folder on GitHubat commit e02ec7e

Compare with similar skills

Revision next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Revision compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Revision this skilljwynia/agent-skills170—~4.8kAutomated safety check: PassMIT
User-Facing Text Cleanupguillaumemeyer/watermarks-remover24k—~3.5kAutomated safety check: PassMIT
Story Multi-Perspective Reviewzenstory-ai/oh-story-claudecode7.4k3 repos~3kAutomated safety check: PassMIT
Chinese Text Humanizerop7418/Humanizer-zh19k—~2kAutomated safety check: PassMIT
Baoyu TranslateJimLiu/baoyu-skills27k1 repos~3.9kAutomated safety check: PassMIT
Natural Japanese Business Writingcoji/natural-japanese1.9k—~2.1kAutomated safety check: PassMIT

Similar skills

  • User-Facing Text Cleanup

    guillaumemeyer/watermarks-remover

    Audits prose for invisible Unicode characters and rewrites it while keeping facts, citations, code and required disclosures unchanged and the writer's voice intact.

    24k GitHub stars~3.5k tokensUpdated 2 days ago
    Writing & ContentAuto-check passed

  • Story Multi-Perspective Review

    zenstory-ai/oh-story-claudecode

    Reviews Chinese web-novel text with several reviewer agents in parallel, falling back to a single-agent pass, and reports structure, character, prose and setting problems with fixes.

    7.4k GitHub starsUsed in 3 repos~3k tokens
    Writing & ContentAuto-check passed

  • Chinese Text Humanizer

    op7418/Humanizer-zh

    Edits Chinese articles, comments and documents to remove filler, repetition and template phrasing while keeping the facts, the level of certainty and the author's voice.

    19k GitHub stars~2k tokensUpdated 17 days ago
    Writing & ContentAuto-check passed

  • Baoyu Translate

    JimLiu/baoyu-skills

    Translates articles and files in quick, normal or refined mode, with a custom glossary, saved preferences and a review-and-polish workflow for publication quality.

    27k GitHub starsUsed in 1 repo~3.9k tokens
    Writing & ContentAuto-check passed

  • Natural Japanese Business Writing

    coji/natural-japanese

    Writes and edits Japanese business documents so they read clearly and naturally, removes AI-sounding phrasing and can score how AI-like a text reads.

    1.9k GitHub stars~2.1k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed

  • Defensive Writing Editor

    lennney/stop-that-shit

    Cuts defensive disclaimers, stacked hedging and self-protective narration from proposals and summaries, keeping only limits that affect the reader's decision.

    2.5k GitHub starsUsed in 1 repo~1.2k tokens
    Writing & ContentAuto-check passed

More from jwynia/agent-skills

All 111 skills in this repo

  • Devcontainer

    jwynia/agent-skills

    Diagnose devcontainer configuration problems and guide development environment setup.

    170 GitHub stars~1.2k tokensUpdated 7 mo ago
    Auto-check: notes

  • Frontend Design

    jwynia/agent-skills

    Create distinctive, production-grade frontend interfaces with high design quality.

    170 GitHub stars~3.2k tokensUpdated 7 mo ago
    Auto-check passed

  • Gitea Workflow

    jwynia/agent-skills

    Orchestrate agile development workflows for Gitea repositories using the tea CLI.

    170 GitHub stars~3.8k tokensUpdated 7 mo ago
    Auto-check passed

  • Godot Asset Generator

    jwynia/agent-skills

    Generate game assets using AI image generation APIs (DALL-E, Replicate, fal.ai) and prepare them for Godot.

    170 GitHub stars~3.8k tokensUpdated 7 mo ago
    Auto-check passed

  • Mastra Hono

    jwynia/agent-skills

    Develop AI agents, tools, and workflows with Mastra v1 Beta and Hono servers.

    170 GitHub stars~2.9k tokensUpdated 7 mo ago
    Auto-check passed

  • PPTX Generator

    jwynia/agent-skills

    Create and manipulate PowerPoint PPTX files programmatically.

    170 GitHub stars~3.1k tokensUpdated 7 mo ago
    Auto-check passed

Categories

Questions about Revision

What does Revision do?

Guide the edit pass after drafting. An agent skill from jwynia/agent-skills. Revision is an agent skill from jwynia/agent-skills. Guide the edit pass after drafting.

When should I use Revision?

Revision fits situations like: revision feels overwhelming; changes cascade unpredictably; you cant see problems anymore; editing never ends.

How do I install Revision in Claude Code?

Run `npx skills add jwynia/agent-skills --skill revision -a claude-code`. Or copy the skill folder (skills/creative/fiction/craft/revision in jwynia/agent-skills) into .claude/skills/revision in your project. Claude Code loads it when a task matches its description.

How do I install Revision in Codex?

Run `npx skills add jwynia/agent-skills --skill revision -a codex`. Or copy the skill folder (skills/creative/fiction/craft/revision in jwynia/agent-skills) into .agents/skills/revision in your project. Codex loads it when a task matches its description.

Can I use Revision in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jwynia/agent-skills --skill revision -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/revision, .gemini/skills/revision, .github/skills/revision and .opencode/skills/revision in your project.

What does Revision need to run?

Going by SKILL.md and its folder, Revision needs TypeScript for the scripts in its folder and the command-line tools its instructions call (deno). Our summary lists: Node.js.

Does Revision access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Revision safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Revision use?

Revision is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Revision use?

About 4.8k tokens (SKILL.md is roughly 19k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Revision?

Skills that share tags, products or a category with Revision: User-Facing Text Cleanup (guillaumemeyer/watermarks-remover, 24k stars), Story Multi-Perspective Review (zenstory-ai/oh-story-claudecode, 7.4k stars), Chinese Text Humanizer (op7418/Humanizer-zh, 19k stars) and Baoyu Translate (JimLiu/baoyu-skills, 27k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Revision?

jwynia (a GitHub user) maintains it in jwynia/agent-skills, which has 170 GitHub stars. The repository holds 111 skills in this directory. The repository was last updated on February 24, 2026.

Source: jwynia/agent-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.