一键在 Manus 中运行任何 Skill

report

星标2

分支0

更新时间2026年6月26日 09:48

This skill should be used when the user invokes /perfmind:report, says "generate a performance report", "write a summary for my manager", "create a performance write-up for leadership", "show me recommendations as a DevOps engineer", "give me actionable fixes as a developer", "write an analyst-level summary", "I need to share these findings with the team", "write a report for my team", "document these perf findings", "share findings with stakeholders", or asks to document performance findings for a specific audience. Reads investigation findings from the conversation and auto-detects the target audience.

安装

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。

在 Manus 中运行

来源

sembraniteam

sembraniteam/claude-plugins

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

相关职业SOC

基于 SOC 职业分类

软件开发工程师计算机与数学类职业·SOC 15-1252

文件资源管理器

2 个文件

SKILL.md

readonly

同仓库更多 Skills

同仓库

parallel-debug

sembraniteam/claude-plugins

This skill should be used when the user invokes /debugging-workflow:parallel-debug, asks to "debug in parallel", "try all hypotheses at once", "spawn multiple agents for this bug", "investigate multiple root causes", "parallel hypothesis testing", "parallel investigation", "I can't figure out what's causing this and have a few theories", "investigate from multiple angles", "I have multiple theories about this bug", or reports a complex or intermittent bug where the root cause is unclear and multiple theories need testing simultaneously.

2026-07-112

design

sembraniteam/claude-plugins

This skill should be used when the user wants to design a new application's architecture or infrastructure — says "design my architecture", "help me plan the architecture", "create architecture diagrams", "I need to plan a new system", or is starting a new project and needs a structured design process. Also trigger when the user mentions HLD, LLD, API contracts, or system design. Guides from requirements gathering through capacity planning, technology selection, diagram generation, low-level design, document saving, and code skeleton implementation.

2026-07-102

implement

sembraniteam/claude-plugins

This skill should be used when the user wants to turn an approved architecture document into working code, or wants to generate/scaffold a project after a design or review session — says "implement the architecture", "scaffold the project", "generate the code from my architecture", "create project files from the design", "turn my architecture into code", "create the folder structure", "start implementation", "let's start coding", or "generate the project skeleton".

2026-07-102

review

sembraniteam/claude-plugins

This skill should be used when the user wants to review or revise an existing architecture, says "review my architecture", "audit my architecture", "check my architecture", "my architecture needs review", "update my architecture document", "revise the architecture", "architecture drift", "compare design vs implementation", "architecture is outdated", "architecture inconsistency", "check if my code matches my design", or wants to compare their architecture document against their current codebase. Also trigger when the user mentions their architecture document needs updating after new features were added or requirements changed.

2026-07-102

generate-branch

sembraniteam/claude-plugins

This skill should be used when the user invokes /git-helper:generate-branch, asks to "create a branch name", "name my branch", "what should I name this branch", "suggest a branch name", "what prefix should I use", "what branch should I create for this", "branch for ticket

2026-07-102

generate-commit

sembraniteam/claude-plugins

This skill should be used when the user invokes /git-helper:generate-commit, or asks to "generate a commit message", "write a commit for me", "what should my commit message be", "help me commit my changes", "suggest a commit message", "prepare a commit", "what commit type should I use", "stage and commit my changes", "commit all my changes", "I'm done working on this", "commit everything I did", "write a commit for my changes", or describes a change they just made and needs a commit message for it. Produces a conventional commit with type, scope, subject, optional body, and breaking change footer.

2026-07-102

name	report
description	This skill should be used when the user invokes /perfmind:report, says "generate a performance report", "write a summary for my manager", "create a performance write-up for leadership", "show me recommendations as a DevOps engineer", "give me actionable fixes as a developer", "write an analyst-level summary", "I need to share these findings with the team", "write a report for my team", "document these perf findings", "share findings with stakeholders", or asks to document performance findings for a specific audience. Reads investigation findings from the conversation and auto-detects the target audience.
argument-hint	[role] e.g. developer \| devops \| leadership \| perf-engineer
allowed-tools	Read

Performance Report

Role-Tailored Performance Report Generation

Generate a structured performance report adapted to the target audience.

Step 1: Gather Findings

Read the performance findings from the current conversation context. If no /perfmind:investigate session has been run, prompt the user to share key findings first or invoke /perfmind:investigate before proceeding.

Step 2: Detect Target Role

If the user passes a role argument, use it directly. Otherwise, infer from conversation context:

Signal words	Role
"deploy", "infra", "Kubernetes", "scaling", "SRE", "runbook", "alert"	DevOps
"code fix", "PR", "refactor", "function", "file", "commit"	Developer
"flame graph", "profiler", "benchmark", "regression", "p99", "percentile"	Perf Engineer
"SLA", "user experience", "cost", "roadmap", "business impact", "budget"	Leadership

If unclear, default to Developer.

Step 3: Generate Role-Specific Report

Generate the report using the full template for the detected role from references/role-templates.md.

Key output principles per role:

Role	Focus	Tone	Length
Developer	Code-level fixes, file:line references, patch-sized items	Technical, specific	No limit
Perf Engineer	Baseline metrics, evidence chain, data gaps, regression prevention	Analytical, metric-first	No limit
DevOps	SLO status, infrastructure actions, runbook entry	Operational, config-focused	No limit
Leadership	Business impact, effort vs. timeline, risk if not addressed	Non-technical, outcome-focused	<400 words

After generating the report, output: "Share more evidence via /perfmind:investigate to refine these recommendations."

Additional Resources

references/role-templates.md — Full markdown templates for all four roles with section-by-section guidance and audience notes