performance-monitor

Email teams can make costly decisions from underpowered or poorly controlled tests. This skill creates a structured experiment plan and interprets results using user-provided campaign data.

Email exports often mix active subscribers, lapsed customers, and people who must not receive messages. This skill converts your own exported data into documented segments and a suppression list with clear confidence labels.

Paid campaigns can lose relevance when the ad promise does not match the landing page. This skill checks the visitor experience and provides a clear pass, partial, or fix list.

Campaign teams need clear controls to prevent ads from appearing in unsuitable placements. This skill creates an evidence-backed exclusion plan and packages the material for an A1 brand-safety review.

Revenue-only conversion values can make value-based bidding optimize for low-margin orders. This skill creates a documented conversion-value model and QA checklist using margin, COGS, and defensible proxy values.

Subscribers may unsubscribe when messages arrive too often or cover unwanted topics. This skill designs clear topic, cadence, and pause choices that preserve a direct full opt-out path.

Scientific datasets often arrive in specialized formats that are hard to inspect quickly. This skill detects the format, summarizes contents, checks quality, and produces a clear EDA report.

Single-cell RNA-seq workflows require careful QC, normalization, clustering, and documentation. This skill gives Scanpy workflows, scripts, templates, and references for reproducible analysis.

Scientific figures often fail journal checks for size, color, typography, or export format. This skill guides Claude, Codex, and Claude Code through publication-ready plotting workflows.

Structured configuration data can be hard to inspect across JSON, YAML, and TOML files. This skill guides jq and yq queries for extraction, filtering, conversion, and updates.

Therapeutic machine learning requires reliable datasets, meaningful splits, and consistent evaluation. This skill guides PyTDC workflows from dataset selection through molecular optimization.

Researchers often need precise data availability text, repository plans, and FAIR checks before submission. This skill turns English or Chinese data notes into Nature-style statements, dataset citation actions, and missing-information flags.