个人作者

yibo

Data Science and Artificial Intelligence

所在地 · Pluto更新于 2026-09-05

Skills
31
仓库
1
职业覆盖
0
场景覆盖
0
Stars
★ 4

yibo 的 Skills 31

yibocat
yibocat/prismnext

demo-multi-file

Local test skill with an extra notes file

★ 4
yibocat
yibocat/prismnext

writing-results

Use when drafting a Results / Experiments / Analysis section — narrating verified evidence from run receipts, with complete statistical reporting, derivation analysis, and honest treatment of negative results. Every number traces to a run; every claim survives the audit.

★ 4
yibocat
yibocat/prismnext

writing-related-work

Use when drafting a Related Work / literature review section or a standalone synthesis note — organizing prior work into a grounded, cited narrative or comparing methods across papers, from the project library plus staged external discovery. Never cites from memory.

★ 4
yibocat
yibocat/prismnext

writing-preliminaries

Use when drafting a Preliminaries / Background / Notation section — definitions, problem setup, and formal statements the rest of the paper builds on, with notation consistency and machine-checked formal claims. Reference patterns, not molds; include only what later sections actually use.

★ 4
yibocat
yibocat/prismnext

writing-methods

Use when drafting a Methods / Methodology / Approach / Model section — expounding the ideas and the reasoning behind the method, grounded in the project's brief, notes, and prior discussions, with empirical details from run receipts. Structure follows the idea's shape; reference angles, not fixed templates.

★ 4
yibocat
yibocat/prismnext

writing-introduction

Use when drafting or rewriting a paper's Introduction — framing the problem, positioning against prior work, and stating contributions as checkable claims, grounded in the brief, notes, and the library. Reference rhetorical patterns, adapted to the paper — never applied as molds.

★ 4
yibocat
yibocat/prismnext

writing-design

Use when starting to write a paper, report, or any substantial document from the project's research — the outline gate. Discuss the story, audience, and section plan with the user and fix the outline on disk BEFORE any section is drafted. Not for revising a single existing section (use the section skill directly).

★ 4
yibocat
yibocat/prismnext

writing-conclusion

Use when drafting a Conclusion / Discussion section — answering the questions the Introduction raised, stating honest limitations, and marking what the evidence does and does not license. No new claims appear here.

★ 4
yibocat
yibocat/prismnext

symbolic-math

Use when verifying a novel or uncertain symbolic claim — a hypothesis, a conjectured identity or equation proposed in discussion, or a derivation step whose correctness is not already established — with SymPy checking and LaTeX output that goes straight into the manuscript. Not for restating established formulas or literature results with known derivations (cite them, don't re-derive), formal theorem proving, or purely numerical computation.

★ 4
yibocat
yibocat/prismnext

statistical-rigor

Use when choosing statistical tests, computing power or sample size, reporting effect sizes, or correcting for multiple comparisons — and when auditing a results section or a claimed finding for statistical validity.

★ 4
yibocat
yibocat/prismnext

skill-creator

Use when the user wants to create, distill, or author a new agent skill — turning a workflow that just worked in conversation into a reusable skill, or authoring one from scratch — including multi-file skills with references, templates, scripts, and assets.

★ 4
yibocat
yibocat/prismnext

rebuttal-letter

Use when responding to peer review — mapping every reviewer point to a manuscript change and drafting a point-by-point response / rebuttal letter.

★ 4
yibocat
yibocat/prismnext

prisma-systematic-review

Use when conducting a systematic review or structured literature survey — protocol, PRISMA flow, screening log, and synthesis, wired to the project's literature tools. A systematic review is a *method* (predefined protocol, auditable screening, reproducible counts), not an ad-hoc reading pile — for casual exploration or a quick coverage pass, this skill is overkill.

★ 4
yibocat
yibocat/prismnext

ml-research-protocol

Use when running empirical machine-learning experiments — baseline and ablation discipline, multi-seed aggregation, reproducibility logging — or when auditing whether a conclusion you rely on is actually supported by the runs behind it. Wires the experiment island to statistical-rigor and the manuscript.

★ 4
yibocat
yibocat/prismnext

math-numeric

Use when verifying a novel or uncertain mathematical claim numerically — matrix or tensor identities too large for symbolic checking, gradient/Jacobian implementations against finite differences, convergence or consistency order of a numerical method, quadrature or ODE-solution residuals — via seeded random probes, worst-case error reporting, and a PASS/FAIL exit code. Not for restating established textbook or literature numerics (cite them), claims that simplify symbolically (symbolic-math), statistical hypothesis testing (statistical-rigor), or training/optimization runs.

★ 4
yibocat
yibocat/prismnext

math-manifold

Use when verifying novel or uncertain differential-geometry claims about a concrete structure — a given metric, connection, chart, Lie group, or local bundle presentation — including Christoffel/Riemann/Ricci identities, metric compatibility and torsion, geodesic and exp/log map checks, parallel transport and holonomy, Euler-Lagrange and action-stationarity residuals, gauge equivariance, curvature of a connection, and symplectic closedness, via symbolic tensor computation in low dimension plus seeded numeric probes with PASS/FAIL exit codes. Not for restating established textbook or literature geometry (cite them), abstract existence/uniqueness/classification theorems, flat Euclidean checks without geometric structure (math-numeric), or non-geometric symbolic identities (symbolic-math).

★ 4
yibocat
yibocat/prismnext

math-lattice

Use when verifying novel or uncertain concrete computations on polynomial rings, ideals, and lattices — Groebner-basis ideal membership and ideal equality, quotient-ring relations, LLL-reduced basis equivalence, lattice vector membership, short-vector bounds, minimal polynomials, norms and units in small algebraic number fields — with SymPy and fpylll scripts producing PASS/FAIL exit codes. Not for restating established textbook or literature algebra (cite them), symbolic identities without ring structure (symbolic-math), numeric matrix checks without lattice structure (math-numeric), or abstract structure theorems (cite them or use a proof assistant).

★ 4
yibocat
yibocat/prismnext

manuscript-preflight

Use before submission, sharing, or tagging a manuscript release — compile, citation integrity, bibliography sync, figure paths, desk-reject and anonymization checks, and git hygiene as one pass/fail checklist. Mechanical integrity only — it does not judge the argument or the science.

★ 4