Agent skills for Microsoft Copilot

Orchestrates the end-to-end academic publishing lifecycle for submissions to Scientia Sinica Informationis from topic selection to…

Use when the user needs to design, review, adapt, or rehearse scientific communication for a live, spoken, visual, or audience-facing…

Optimizes academic manuscripts for the Scientific Decision Making journal by aligning framing, methodology, and formatting with its…

Use only when the current deliverable is holistic, prioritized, multi-lens critique of audience-facing scientific communication or…

Optimizes academic manuscripts for submission to the Scientific Management Research journal, focusing on innovation policy and S&T…

This is the core skill for the deep research and writing tool—combining AI-driven deep research with well-formatted written outputs.

Machine learning: clustering, PCA/t-SNE/UMAP, classification, prediction regression (Ridge/Lasso/ensemble), cross-validation, Pipelines.

Scikit-learn model training skill with cross-validation, hyperparameter tuning, pipeline construction, and model serialization.

When to activate: scikit-learn, sklearn, Pipeline, ColumnTransformer, GridSearchCV, cross-validation, custom transformers, feature…

Use when the user has a clear-enough task and wants it decomposed into batched worker sub-tasks before any code is written.

Design a scope and sequence showing vertical and horizontal curriculum coherence across a programme or year.

Verify whether a proposed feature or task is within the ManyMoons v1 GDD scope.

Detects scope creep by quantifying drift percentage.

Judgment check before calling substantial work done or proposing a commit.

Load at the start of any build, fix, refactor, or "make it better" request, and again before every follow-up iteration on work already…

Design and implement scoped permission models

Use before any writing begins to stress-test whether a result clears PNAS's bar — high quality and broad significance to a general…

Use before any writing — as the first gate — to stress-test whether a result fits PNAS Nexus — high-quality, broadly significant…

Use before any writing to stress-test whether a result clears Science's (AAAS) desk-reject filter — broad significance and general…

Full card news generation pipeline — auto-selects sequential (subagent) or collaborative (Agent Teams) mode based on environment

Quality assurance review with JURI (ethics) and MARU (empathy) — read-only expert review of generated card news