seanrobertwright
seanrobertwright/lril-skills
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/seanrobertwright/lril-skills --skill handoff命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Build a complete User Acceptance Test for a codebase — an exhaustive markdown checklist written to the tester's skill level (novice / intermediate / expert), plus an HTML form the tester fills in (pass/fail/not done, look-and-feel comments, screenshots) that writes their answers back into the markdown. Use when the user asks for a UAT, an acceptance test checklist, a manual test plan, a "test everything" script for a non-technical tester or a QA engineer, or wants someone to sign off on an app before release.
Turn a completed UAT into a prioritised fix plan and then fix it. Reads a UAT markdown file (or its results.json) that a tester has filled in, diagnoses every failure, look-and-feel complaint and tester-reported finding against the actual code, and produces a plan grouped by severity. Use when a UAT has been submitted, when the user says "process the UAT", "the tester finished", "here are the UAT results", or points at a file with failed acceptance tests.
Audit and prune CLAUDE.md and other instruction files using progressive disclosure — move situational content out of every-session context into path-scoped rules, skills, nested CLAUDE.md files, or referenced docs, and delete what the codebase already tells Claude. Use when a CLAUDE.md is long or bloated, when startup context is too full, when instructions are being ignored, when asked to prune, trim, slim, shrink, audit, or restructure CLAUDE.md / project instructions / memory files, or when the user says "context diet", "my CLAUDE.md is too big", or asks to reorganize instructions into .claude/rules or .claude/references.
基于 SOC 职业分类
正在显示 SKILL.md
| name | handoff |
| description | Compact the current conversation into a handoff document for another agent to pick up. |
| argument-hint | What will the next session be used for? |
| disable-model-invocation | true |
Write a handoff document summarising the current conversation so a fresh agent can continue the work. Save to the temporary directory of the user's OS - not the current workspace.
Include a "suggested skills" section in the document, which suggests skills that the agent should invoke.
Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead.
Redact any sensitive information, such as API keys, passwords, or personally identifiable information.
If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly.