x-experimental-ops

x-experimental-ops is a skill for Claude Code from ElemontCapital/x-algorithm-skills. It costs 38 tokens per session (539 once invoked), scanned A, original, Apache-2.0.

A guide to X's internal experiments and the measures used to judge changes to its recommendation algorithm. A/B testing compares different versions with separate groups of users.

In plain words
What is it for?
Use it to reason about DuckDuckGoose experiment assignments, rollout groups, sticky variants, Unregretted User Minutes, and changes to machine-learning components.
Why use it?
It helps explain why users may see different behavior and how experiments, user groups, and success metrics affect decisions.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the x-algorithm plugin — 8 skills shipped together

Good fit Use it to reason about DuckDuckGoose experiment assignments, rollout groups, sticky variants, Unregretted User Minutes, and changes to machine-learning components.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/elemontcapital/x-algorithm-skills/x-experimental-ops
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add ElemontCapital/x-algorithm-skills --skill x-experimental-ops
Clone the repo
git clone --depth 1 https://github.com/ElemontCapital/x-algorithm-skills

Made for: Claude Code.

Or install x-algorithm, the plugin that ships this one along with the rest of its 8 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for x-experimental-ops

README.md
[![agentmods](https://agentmods.dev/badge/skills/elemontcapital/x-algorithm-skills/x-experimental-ops.svg)](https://agentmods.dev/skills/elemontcapital/x-algorithm-skills/x-experimental-ops)
Your own site
<a href="https://agentmods.dev/skills/elemontcapital/x-algorithm-skills/x-experimental-ops"><img src="https://agentmods.dev/badge/skills/elemontcapital/x-algorithm-skills/x-experimental-ops.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 539 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00038 $0.00539
Opus 5 $0.00019 $0.00269
Sonnet 5 $0.00008 $0.00108
Haiku 4.5 $0.00004 $0.00054

Measured 8d ago against content hash 5eab31d8d8b8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

x-experimental-ops scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/x-algorithm/skills/x-experimental-ops/SKILL.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

X Experimental Ops

Knowledge of X's A/B testing infrastructure (DuckDuckGoose) and the metrics used to measure algorithmic success.

Context

The algorithm is never "finished." It is a living system managed by DuckDuckGoose (DDG), X's internal experimentation platform. Every change to a weight or a filter is first tested on a small percentage of the user base.

What it does

  • Explains Bucketing:
    • Details the mechanics of DuckDuckGoose, X's internal experimentation platform that uses salt-based consistent hashing to deterministically assign users to "Control" or "Treatment" variants.
    • Ensures "sticky" assignments so a user's experience remains consistent across sessions while maintaining statistically sound percentage-based rollouts (e.g., 1%, 5%, or 10% cohorts).
  • Decodes Success Metrics:
    • Breaks down the "Unregretted User Minutes" (UUM) North Star metric, which prioritizes high-value time spent (replies, likes, and deep reads) over passive scrolling or "clickbait" interactions that lead to user regret.
    • Analyzes how experimental changes impact the Multi-Task Learning (MTL) "heads" to ensure a boost in one engagement signal (like Retweets) doesn't negatively correlate with platform health or retention.
  • Analyzes Feature Flags:
    • Identifies how the system uses Dynamic Configuration and Feature Gates to toggle ranking logic or retrieval sources on and off for specific cohorts in real-time.
    • Explains the "Kill Switch" architecture that allows engineers to instantly roll back a new algorithmic feature if it causes a spike in latency or negative feedback without requiring a full code redeployment.

Example Trigger Prompts

  • "/run-experiment salt-based hashing for user buckets"
  • "/run-experiment Unregretted User Minutes vs dwell time"
  • "Trace feature flag logic for latest Grok retrieval test"
  • "Show holdout group parameters for current Heavy Ranker A/B"
  • "Compare control vs variant metrics for feed engagement test"
  • "Explain how a new signal is staged in an experiment pipeline"

Read the full file on GitHub · 37 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 37 lines · 38 tokens per session scan A 5eab31d8d8b8

Subscribe to this mod's changes

x-experimental-ops is a skill published in the GitHub repository ElemontCapital/x-algorithm-skills (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 38 tokens to every session and 539 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.