X algorithm
Use when creating, reviewing, or improving X posts with guidance grounded in the public X For You algorithm release, or when working with the X For You feed stack including Home Mixer, Phoenix retrieval/ranking, Thunder, Grox content understanding, ads, filters, hydrators, scorers, and recommendation-system inference artifacts.From its SKILL.md
npx -y skills add ChloeVPin/x-algorithm-skill-kit --skill x-algorithmAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.5 KB, 435 tokens by cl100k_base, as published. Nobody here has run it
X Algorithm
How To Use This Skill
This skill is self-contained. Use the bundled references when the original X algorithm repository is not installed.
For post creation or review, load references/post-strategy.md first. Treat it as evidence-grounded guidance, not a promise of reach. Default to direct, specific, ready-to-post writing that matches the user's requested voice. Do not claim exact production weights, private ranking rules, or guaranteed distribution.
For engineering work, use the architecture references to understand, explain, or modify the X For You recommendation stack: Home Mixer orchestration, Phoenix retrieval/ranking, Thunder in-network serving, Grox content-understanding tasks, candidate pipelines, ads, filters, hydrators, scorers, selectors, or side effects.
Reference Map
- Post creation and review guidance: references/post-strategy.md
- Full architecture and workflow: references/overview.md
- Pipeline execution model and stage ordering: references/pipeline.md
- Subsystem details for Home Mixer, Phoenix, Thunder, Grox, and ads: references/subsystems.md
- Validation, test, and modification rules: references/operations.md
Working Rules For Posts
- Optimize for predicted user value: likely replies, reposts, clicks, dwell, shares, follows, and low negative feedback.
- Preserve eligibility: avoid spam patterns, bait, misleading framing, unsafe content, muted-keyword triggers, duplicate posts, and low-quality replies.
- Build posts around a clear audience and topic so retrieval, embeddings, and user-history matching have strong signals.
- Prefer specific, self-contained posts over vague engagement prompts.
- When reviewing a draft, return concrete edits and a short rationale tied to the bundled algorithm references.
- For drafting, return the ready-to-post copy first, then optional notes only when useful.
If The Repo Is Present
If the workspace also contains the repo, use rg first and then inspect the narrowest file set for the requested task.
What ships with it: 7 files
25.0 KB alongside SKILL.md
agents/
- openai.yaml287 B
references/
- operations.md2.1 KB
- overview.md4.1 KB
- pipeline.md2.2 KB
- post-strategy.md10.6 KB
- subsystems.md4.7 KB
- LICENSE.txt1.0 KB