agentleFS
AirMileAirMile/claude-config/skills/content-rewrite/SKILL.md

content-rewrite

Skill0 starsChanged 4 months agoMITView on GitHub

Rewrite text in a configured writing style. Use with /content-rewrite.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    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. Very specific:

    If first token equals check: set mode = "check", shift it off.

    line 28
  2. Very specific:

    If no styles are found, output: "No styles configured.

    line 36
  3. Very specific:

    If style is null, ask (present the discovered style names as options, using the first non-header line of each style file as its …

    line 40
  4. Very specific:

    If text is null, ask the user to provide it (no fixed format — accept whatever they paste or reference via @path).

    line 52
  5. Very specific:

    Do NOT rewrite. Walk through text sentence-by-sentence and check each rule from the loaded style profile.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Rewrite

Rewrite freewritten text in a chosen writing style, or check text for style compliance.

1. Parse Input

The argument can take these forms:

  • [style] [text] — rewrite text
  • [style] @path/to/file.md — read file, rewrite content
  • check [style] [text|@path] — compliance check, no rewrite
  • [text] — ask which style, then rewrite
  • (empty) — ask for both style and text

Detection order:

  1. If first token equals check: set mode = "check", shift it off. Otherwise mode = "rewrite".
  2. Discover available styles:
    # Primary: user-owned styles
    ls ~/.claude/styles/*.md 2>/dev/null | xargs -I{} basename {} .md
    # Fallback if directory is empty or doesn't exist:
    ls "$CONFIG_REPO/skills/shared/styles/style-*.md" 2>/dev/null | xargs -I{} basename {} .md | sed 's/^style-//'
    If no styles are found, output: "No styles configured. Add style files to ~/.claude/styles/ — see skills/shared/styles/style-example.md for the format." Then stop.
  3. Next token: match against discovered style names. On match, set style and shift it off. Otherwise style = null.
  4. Remaining argument: if it starts with @, read the file at that path and use its content as text. Otherwise treat the remainder as text.

If style is null, ask (present the discovered style names as options, using the first non-header line of each style file as its description):

header: "Style"
question: "Which writing style?"
options:
  - label: "{style-name}"
    description: "{first non-header line from the style file}"
  # … one entry per discovered style
multiSelect: false

If text is null, ask the user to provide it (no fixed format — accept whatever they paste or reference via @path).

2. Load Style Profile

Load the style file for the selected style:

# Primary: user-owned styles
STYLE_FILE="$HOME/.claude/styles/style-{style}.md"
# Fallback:
STYLE_FILE="$CONFIG_REPO/skills/shared/styles/style-{style}.md"
Read(resolved STYLE_FILE path)

Apply all rules from the loaded style file strictly. The style file is the single source of truth — do not blend styles or add rules not in the profile.

3. Execute

If mode == "rewrite"

Apply every rule from the loaded style profile to text. Follow the profile strictly. Do not blend styles or add rules not in the profile.

Preserve:

  • The original meaning and information
  • Technical terms and proper nouns
  • The language of the input (Dutch stays Dutch, English stays English)

If mode == "check"

Do NOT rewrite. Walk through text sentence-by-sentence and check each rule from the loaded style profile. Output format:

Style: {style}

✓ {rule that passes}: {brief evidence}
✗ Zin {N}: "{quote}" — {rule violated, brief reason}
...

Score: {X overtredingen / Y zinnen}

Group passes at the top, violations below, ordered by sentence number. Quote the offending text inline. Keep each violation to one line — name the rule, name what's wrong, no over-explanation.

4. Output

Rewrite mode

Output ONLY the rewritten text. No explanations, no "here's the rewrite", no meta-commentary.

Check mode

Output ONLY the check report (format above). No rewrite, no fix-suggestions unless explicitly asked in a follow-up.

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.