SituationalMITOriginal

Ghostwrite In Voice

Drafts in a specific author's voice after reading 3-5 of their pieces — with attribution-safe output for transparent ghostwriting.

Entry verified April 21, 2026

The short answer

For agencies, ghostwriters and internal comms teams. The skill reads sample pieces, extracts a voice brief, and drafts new content that reads like the named author wrote it. The ethical use is stated-up-front ghostwriting (client byline, consented) — not deception.

When to use it

Client ghostwriting. Executive comms and founder-voice pieces. Translating between a brand's founder voice and a new team member's draft so the house voice stays consistent.

Setup

  1. 1

    Save the file as ~/.claude/commands/ghostwrite.md.

  2. 2

    Feed sample pieces from the target author; skill produces a voice brief plus a draft in that voice.

  3. 3

    Add disclosure language to the output — e.g. "draft ghostwritten per client brief" — so the chain of authorship stays clear.

  4. 4

    Use only where ghostwriting is known and consented; never for impersonation.

Example

You: /ghostwrite — attached: 3 Medium posts. Target: 1,200-word piece on vibe coding.
Claude: [analyses voice → writes draft that matches cadence, vocab, punctuation habits]

Source & attribution

Author
Bryan Collins (adapted from brand-voice pattern)
Licence
MIT
Type
Original

Original pattern published under MIT — attribution preserved by convention, not licence requirement.

Caveats

Ethical concerns — use only where ghostwriting is known and consented. Never for impersonation.

New skills, explained plainly.

One short email when a new Claude Code skill is worth installing — and when to skip the hype. No spam.

56 skills across 10 categories, all licence-verified.