x-voice-match_skill

This skill analyzes a Twitter/X account to capture posting patterns and generates authentic posts that match the owner's voice.
  • Python

2.6k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill x-voice-match

  • _meta.json280 B
  • SKILL.md3.5 KB

Overview

This skill analyzes an X (Twitter) account’s posting patterns and generates posts that match the account owner’s unique voice. It produces a structured voice profile and style-matched post suggestions, including confidence scores and reasoning. Use it to maintain consistent voice, draft tweets quickly, or study posting behavior over time.

How this skill works

The skill ingests a sample of recent tweets or a tweet export and extracts measurable patterns: length distributions, tone markers, topic focus, engagement style, signature phrases, emoji and punctuation habits, and content types. It builds a JSON voice profile and uses that profile to generate 1–N candidate posts for a given topic or post type, annotating each candidate with a confidence score and a short justification.

When to use it

  • Create X posts that sound like a specific account
  • Maintain consistent voice across multiple posts or accounts
  • Analyze posting patterns for content strategy or archival research
  • Generate batches of topic-aligned posts in the target voice
  • Recreate voice for accounts with public archives or backups

Best practices

  • Analyze 30–50 recent tweets for a reliable profile; 50+ increases accuracy
  • Prefer recent tweets to capture current tone and topics
  • Limit generation to topics the account commonly covers for higher authenticity
  • Review and edit generated posts before publishing; confidence <70% may need iteration
  • Re-run analysis periodically to capture voice evolution

Example use cases

  • Produce 3–5 style-matched tweet options for a conference thread
  • Audit an account’s engagement patterns to inform posting schedule and CTAs
  • Batch-generate themed posts (crypto takes, memes, observations) that match voice
  • Recover a consistent posting style from an archived account for backup or migration
  • Create reaction posts and hot takes aligned with the account’s typical phrasing

FAQ

Aim for 30–50 tweets as a minimum; 50+ gives more reliable length, tone, and topic patterns.

Can I generate tweets for a topic the account never covered?

You can, but authenticity drops. Results are best when the topic aligns with the account’s historical topics.

What does the confidence score mean?

Scores reflect how closely generated text matches measured patterns (tone, length, phrases, punctuation). Below 70% indicates the output may need revision.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational