在 Manus 中运行任何 Skill
一键导入
一键导入
一键在 Manus 中运行任何 Skill
开始使用$pwd:
$ git log --oneline --stat
stars:197
forks:27
updated:2026年5月22日 18:23
SKILL.md
readonly
$ install --global
$ download --local
$ useful --forSOC
Create and send Gmail emails using draft-first workflow. Always draft first, get user approval, then send.
Systematically debug issues using graph-powered code navigation
Navigate and understand codebase structure using the knowledge graph
Perform a structured code review using change detection and impact
Design world-class product onboarding for SaaS applications that drives activation, reduces churn, and converts users. Use this skill whenever someone mentions onboarding, user activation, first-time user experience, aha moment, time-to-value, user signup flow, welcome experience, product tour, user activation funnel, trial conversion, or reducing churn through better first experiences. Also trigger when someone asks about empty states, getting-started flows, welcome screens, product walkthroughs, or improving new user retention. Even if they say "our users don't get it" or "people sign up but never come back" — that's an onboarding problem and this skill applies.
Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity ratings and actionable plan. Use when the user wants an accessibility check, performance audit, or technical quality review.
| name | Refactor Safely |
| description | Plan and execute safe refactoring using dependency analysis |
Use the knowledge graph to plan and execute refactoring with confidence.
refactor_tool with mode="suggest" for community-driven refactoring suggestions.refactor_tool with mode="dead_code" to find unreferenced code.refactor_tool with mode="rename" to preview all affected locations.apply_refactor_tool with the refactor_id to apply renames.detect_changes to verify the refactoring impact.get_impact_radius before major refactors.get_affected_flows to ensure no critical paths are broken.find_large_functions to identify decomposition targets.