se-uhd

LLM Guidelines Skill — AI skill for Claude Code

AI community

Skill to explore the LLM Guidelines for SE or check papers and supplementary material against them.

How to install LLM Guidelines Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open se-uhd/llm-guidelines-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What LLM Guidelines Skill does

Skill to explore the LLM Guidelines for SE or check papers and supplementary material against them.

Alternatives in AI

Wiki Lint — Health-check the LLM Wiki for issues 2.3k ★
Oil UI — Push AI UI design to its limits: explore distinct directions, compare them side by side, and refine against re 861 ★
onWatch — Open-source Go CLI that tracks AI API quota usage across 7 providers (Synthetic, Z.ai, Anthropic, Codex, GitHu 397 ★

README

LLM Guidelines Skill

An [Agent Skill](https://agentskills.io/home) bundle that applies the community guidelines for empirical software engineering studies involving LLMs to a paper draft and its supplementary material. Agent Skills is an open standard originally developed by Anthropic. The format is now read by Cursor, GitHub Copilot, OpenAI Codex, Gemini CLI, Claude Code, and JetBrains Junie, among others; see the [client list](https://agentskills.io/clients).

The bundle exposes a single skill, `llm-guidelines`, with two modes: explore mode for discussing the guidelines and planning a study, and review mode for assessing a concrete draft against the eight guidelines and writing a report. Claude Code adds slash-command entry points for those modes; Codex CLI invokes the same skill from natural-language prompts.

The guideline content is published at and authored in the companion paper repository . The Markdown files in this repository are generated from those LaTeX sources; see for the conversion pipeline. **Do not edit the generated files in this repository directly.**

Versioning

The bundle uses two-part versioning: a guideline version (CalVer `YYYY.MM`, matching the paper and website tag) and an optional skill revision suffix.

Initial release for a guideline version: bare YYYY.MM (e.g., 2026.05).
Skill-only revisions for the same guideline version: YYYY.MM_revN (e.g., 2026.05_rev1, 2026.05_rev2).

The current Claude/skill version is recorded in `VERSION` and embedded in `.claude-plugin/marketplace.json`, `plugins/llm-guidelines/.claude-plugin/plugin.json`, and `plugins/llm-guidelines/skills/llm-guidelines/SKILL.md` (under `metadata.version`). Codex plugin manifests require strict semver, so `plugins/llm-guidelines/.codex-plugin/plugin.json` maps the same release to `YYYY.M.REVISION` (for example, `2026.05_rev18` be