メニュー
Transform raw ideas into Balaji-style X posts — assertive, numbered, evidence-backed. USE WHEN tweet, X post, Twitter, draft post, publish, social media, process input, score tweet, check quality, generate image, update algorithm, daily digest.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
When the user wants to plan, design, or implement an A/B test or experiment. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," or "hypothesis." For tracking implementation, see AnalyticsTracking. USE WHEN ab test, split test, experiment setup, A/B test, multivariate.
When the user wants to set up, improve, or audit analytics tracking and measurement. Also use when the user mentions "set up tracking," "GA4," "Google Analytics," "conversion tracking," "event tracking," "UTM parameters," "tag manager," "GTM," "analytics implementation," or "tracking plan." For A/B test measurement, see AbTestSetup. USE WHEN analytics, tracking, GA4, measurement, Google Analytics.
Structure raw ideas into articulate outlines using Dan Koe's frameworks. USE WHEN articulate, structure ideas, outline talk, outline article, think through piece, brainstorm structure, how should I say this, organize my thoughts, structure this post, help me outline.
Balaji Srinivasan slide design system. USE WHEN balaji slides, balaji style, evidence slides, balaji presentation, split layout slides, create balaji deck, artifact-driven slides.
When the user wants to create competitor comparison or alternative pages for SEO and sales enablement. Also use when the user mentions 'alternative page,' 'vs page,' 'competitor comparison,' 'comparison page,' '[Product] vs [Product],' '[Product] alternative,' or 'competitive landing pages.' USE WHEN competitor, alternatives page, vs page, comparison.
When the user wants to edit, review, or improve existing marketing copy. Also use when the user mentions 'edit this copy,' 'review my copy,' 'copy feedback,' 'proofread,' 'polish this,' 'make this better,' or 'copy sweep.' USE WHEN edit copy, proofread, copy review, grammar check.
| name | XPoster |
| description | Transform raw ideas into Balaji-style X posts — assertive, numbered, evidence-backed. USE WHEN tweet, X post, Twitter, draft post, publish, social media, process input, score tweet, check quality, generate image, update algorithm, daily digest. |
Autonomous agent for processing raw ideas into Balaji Srinivasan-style X posts. Assertive voice. Numbered points. Evidence in every claim. No engagement bait.
Two formats: Short numbered post (100-500 words) OR long article (500-3000+ words). Nothing in between.
Before executing, check for user customizations at:
~/.claude/skills/PAI/USER/SKILLCUSTOMIZATIONS/XPoster/
When executing a workflow, do BOTH:
Send voice notification:
curl -s -X POST http://localhost:8888/notify \
-H "Content-Type: application/json" \
-d '{"message": "Running the WORKFLOWNAME workflow in the XPoster skill"}' \
> /dev/null 2>&1 &
Output text notification:
Running the **WorkflowName** workflow in the **XPoster** skill...
| Workflow | Trigger | File |
|---|---|---|
| Draft | "/post", "/capture", "draft:", "tweet about", "post about", "turn this into a post" | Workflows/Draft.md |
| BatchProcess | "process input", "scan input", "process captures", "session posts", "batch process" | Workflows/BatchProcess.md |
| Validate | "validate", "score", "quality check" | Workflows/Validate.md |
| Organize | "organize", "move to ready/drafts" | Workflows/Organize.md |
| Schedule | "schedule", "queue for posting" | Workflows/Schedule.md |
| GenerateVisual | "generate image", "add visual" | Workflows/GenerateVisual.md |
| UpdateAlgorithm | "update algorithm", "fetch nikita" | Workflows/UpdateAlgorithm.md |
| DailyDigest | "daily digest", "what did I build today", "session digest", "digest my day", "today's sessions" | Workflows/DailyDigest.md |
bun run ~/.claude/skills/XPoster/Tools/ScoreTweet.ts "Your tweet"
NOTE: ScoreTweet.ts uses legacy scoring logic. Use
ContentRules.mdas ground truth for quality assessment until the tool is updated.
mcp__x-server__create_tweet({ text: "content" })
mcp__x-server__reply_to_tweet({ tweet_id: "id", text: "reply" })
ContentRules.mdXAlgorithm.mdTweetFormats.mdAccountStructure.mdExample 1: Draft from conversation
User: "/post"
-> Invokes Draft workflow (conversation source)
-> Finds thesis in conversation
-> Writes numbered post with evidence
-> Returns Balaji-style draft
Example 2: Draft from idea
User: "draft: replaced Calendly with 200 lines of TypeScript"
-> Invokes Draft workflow (inline source)
-> Structures as numbered post: thesis, evidence points, conclusion
-> Returns assertive, evidence-backed draft
Example 3: Batch processing
User: "process input for software30"
-> Invokes BatchProcess workflow
-> Scans INPUT folder, finds thesis for each item
-> Structures as Balaji-style posts, saves to OUTPUT/drafts/