ワンクリックで
write-method-docs
Write or revise scientific method documentation with equations, algorithm sketches, and implementation-status notes.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Write or revise scientific method documentation with equations, algorithm sketches, and implementation-status notes.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Write or review docstrings for code in this repo. Use when writing new docstrings, reviewing existing docstrings, or editing docstrings.
Polishes recently written or modified Python code for style, dead code, docstrings, type annotations, and checks. Use when coding task is completed.
Write or review code in this repo. Use when writing new code, reviewing existing code, or editing code.
Write or review unit tests for code in this repo. Use when writing new tests, reviewing existing tests, or adding test coverage.
Write or review code in this repo. Use when writing new code, reviewing existing code, or editing code.
| name | write-method-docs |
| description | Write or revise scientific method documentation with equations, algorithm sketches, and implementation-status notes. |
| argument-hint | [topic or target doc path] |
Use this skill when writing theory-first scientific documentation for methods, algorithms, and techniques in this repo.
The goal of these documents is to explain how a method works, why it is used, what equations govern it, how the algorithm proceeds, and how it relates to the current package implementation. The main body should teach the method itself, not just describe code structure.
Use this skill for:
Do not use this skill for:
Most method documents should follow this structure, adapted as needed to the topic:
# <Title>
Introductory paragraph.
## Overview
## Core Theory
## Key Equations
## Algorithm Sketch
## Practical Considerations
## Relationship To Other Methods
## Current Implementation
## References
Not every document needs every heading literally, but each document should contain the following components:
Examples:
# Rational Function Optimization
## Key Equations
### Restricted-step variant
Good candidates for displayed equations include:
Example:
$$
m(s) = g^T s + \frac{1}{2} s^T H s
$$
1. build the local model
2. compute a trial step
3. apply any trust-region restriction
4. evaluate the new point
5. update the model
Place the implementation discussion near the end of the document, after the main scientific content.
This section should answer:
Keep this section distinct from the theory narrative. Do not let the main body collapse into a code tour unless the user explicitly asked for an implementation-focused document.
Every document must end with a ## References section.
Reference rules:
Example style:
## References
1. [Bakken, V.; Helgaker, T. The efficient optimization of molecular geometries using redundant internal coordinates. J. Chem. Phys. 117, 9160–9174](https://doi.org/10.1063/1.1515483)
2. [Baker, J.; Kessi, A.; Delley, B. The generation and use of delocalized internal coordinates in geometry optimization. J. Chem. Phys. 105, 192–212](https://doi.org/10.1063/1.471864)
Good:
See [Trust-Region Newton](trust-region-newton.md) and
[Hessian updates](hessian-updates.md).
Avoid:
See `docs/step-trust-region-newton.md`.
When writing a new scientific method page:
## References.