Skip to main content

code-review

Review a pull request in XRPL-Standards. Use for any PR in this repository, especially one that changes an XLS-*/README.md, the templates, or the validation scripts.

Datos de origen

Repositorio
XRPLF/XRPL-Standards
Última actividad en el origen
11 de septiembre de 2026 a las 20:05
Idioma detectado de SKILL.md
inglés
Estrellas
288
Forks
1567

Opciones de instalación

De forma predeterminada está seleccionado el prompt que primero revisa el origen. Puedes cambiar a un comando directo o descargar una copia local.

Revisa los archivos de origen

Lee SKILL.md y los archivos complementarios que muestra SkillsMP antes de decidir si quieres instalarlo.

Mostrando SKILL.md

SKILL.md
Instrucciones de origen · Vista previa de solo lectura
name
code-review
description
Review a pull request in XRPL-Standards. Use for any PR in this repository, especially one that changes an XLS-*/README.md, the templates, or the validation scripts.
# Reviewing a PR in XRPL-Standards This repository holds prose specifications, not shipping code. A review's job is to find **spec defects**: ambiguity, internal inconsistency, missing normative detail, unsafe design, or a claim that contradicts the implementation. ## Complete the review before commenting Analyze the complete PR before submitting any comments: 1. Read the PR description and complete diff. 2. Inspect every changed file and the unchanged context needed to validate it. 3. For each changed normative behavior, check the affected field tables, failure conditions, state changes, invariants, formulas, examples, RPCs, parent specification, and amendment patches. 4. Build the complete candidate finding set privately. 5. Verify each candidate, remove duplicates and previously addressed findings, then submit one review for the current HEAD commit. Do not publish findings incrementally as they are discovered. Do not stop after finding the first valid issue. ## Follow the repository guidelines Read [`.github/copilot-instructions.md`](../../copilot-instructions.md) — the "Review Guidelines (for AI reviewers)" section is normative for this review. It covers what CI already owns, the process checks, evidence standards, and comment etiquette. ## For any changed `XLS-*/README.md` Follow [`.agents/skills/xls-template-conformity/SKILL.md`](../../../.agents/skills/xls-template-conformity/SKILL.md), including its `references/beyond-the-template.md`. It is the same procedure human contributors run, and it reads `templates/` at review time, so it cannot drift from them. ## Never flag Trailing whitespace, line endings, missing EOF newline, markdown or table formatting (prettier owns table alignment — never suggest realigning a table), missing preamble fields, missing required sections, missing Amendment subsections, leftover template placeholders. Every one of these fails CI on its own. A comment about them is pure noise. Also suppress spelling, grammar, phrasing, verbosity, naming consistency, field qualification, and optional clarity suggestions unless they create two plausible protocol interpretations or would cause incorrect implementation or client behavior. ## Avoid repeat rounds Read existing review threads before submitting the review. Do not repeat an issue that was already reported, resolved, answered, or made outdated. Reopen it only if the same defect still exists at the current HEAD, and explain specifically why the earlier response or change did not resolve it. ## Note on symlinks `.ai-review/instructions.md` is a committed symlink (git mode `120000`). If your view of the tree does not resolve it, it will look absent or look like a one-line file containing a path. Do not report it as missing or malformed. The `.claude/skills/<name>/SKILL.md` files are deliberate stubs: `name` and `description` frontmatter, which Claude Code needs to discover the skill, plus one line pointing at `.agents/skills/<name>/SKILL.md`, which holds the procedure. Do not report them as incomplete, and do not propose removing the frontmatter.
Ver en GitHub