Token导航 LogoToken导航TokenDH.com
待分类只读github未标认证来源可访问许可证需确认审计通过

handshake-acknowledgement握手确认

Agent Skill

handshake-acknowledgement 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

792

周安装

33

GitHub Stars

3

下载量

264
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:handshake-acknowledgement(握手确认)
来源仓库:https://github.com/somnio-software/somnio-ai-tools
仓库路径:skills/handshake-acknowledgement
安装命令:
npx skills add https://github.com/somnio-software/somnio-ai-tools --skill handshake-acknowledgement
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/somnio-software/somnio-ai-tools --skill handshake-acknowledgement

简介

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装使用。
  • 建议确认权限范围和维护状态,注意是否涉及联网或文件操作。
  • handshake-acknowledgement 属于待分类类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

HandShake Acknowledgement Generator

Generates the Somnio HandShake - Step 3 Acknowledgement PDF from raw EM notes. Handles single or batch generation (1 to N devs in one run).

Context

Somnio Software runs a bi-annual performance review program called HandShake (April and October/November). After conducting interviews with the PM, a dev from the team, and the Tech Lead, the EM synthesizes everything into the Acknowledgement document — a formal PDF delivered to the developer.

Input

One or more ficha files (PDF, DOCX, or pasted text) per developer, containing:

  • Notes from PM interview
  • Notes from dev team member interview
  • Notes from Tech Lead review
  • EM synthesis / general feedback
  • Optionally: previous Acknowledgement PDF (for comparison section)

Input is always free-form — Claude must interpret it intelligently.

Output

One PDF per developer, matching the Somnio Acknowledgement template:

  • Somnio logo top-right header, gradient background, transparent footer
  • Header table: Somnier, Líder, Fecha, Rol del Somnier
  • Puntos abordados (standard agenda — always the same)
  • Resumen del encuentro: Seniority · Rendimiento actual · Comparación anterior (if applicable)
  • Objetivos acordados: Oportunidad de mejora + Continuar trabajando en
  • Aclaración o comentarios generales (optional)

Batch Workflow (ALWAYS follow this order)

Step 1 — Read all attached files

When the EM attaches one or more ficha files, read all of them before asking anything. For each file, extract as much as possible:

  • Developer name (Somnier)
  • EM / Líder name
  • Role
  • Current seniority level
  • Target level
  • All feedback content

Step 2 — Show confirmation table

Present a single summary table with one row per developer. Pre-fill every field Claude could infer. Mark fields Claude could NOT determine with [?].

Example format:

Encontré X fichas. Antes de generar, confirmá o corregí los datos:

| # | Somnier | Líder | Fecha | Rol |
|---|---------|-------|-------|-----|
| 1 | John Doe | Jane Smith | [?] | Sr Flutter Developer |
| 2 | Bob Johnson | [?] | [?] | Sr Backend Engineer |

Completá los [?] y avisame si hay algo para corregir. Una vez confirmado, genero todos los PDFs.

Date format rule: Fecha must be Dec 19, 2025 style (abbreviated English month, no leading zero, 4-digit year). Accept any format from the EM and normalise automatically.

Do NOT generate any PDF until the EM explicitly confirms the table.

Step 3 — Generate all PDFs

Once the EM confirms (even partially, e.g. "el resto está bien"), generate all PDFs sequentially. For each dev:

  1. Build the content JSON from the ficha notes
  2. Run the generation script
  3. Report progress: "✅ 1/3 — John Doe.pdf generado"

Deliver all files at the end together.

Step 4 — Deliver

Present all generated PDFs at once using present_files. Name each file: Handshake_Acknowledgement_Firstname_Lastname.pdf


Content Synthesis Guide

From the raw notes, synthesize per developer:

Rendimiento actual

4–7 bullet points highlighting strengths and achievements. Extract from PM/dev/TL feedback. Reformulate cleanly — do not copy raw notes verbatim.

Comparación reunión anterior

Only if a previous Acknowledgement PDF was provided. For each prior objective, assign:

  • Logrado — clearly achieved
  • Logrado, pero hay que seguir trabajando — partial
  • En progreso — started but not there yet
  • Pendiente — not yet addressed

If no previous eval: omit this section entirely.

Oportunidad de mejora

3–6 specific, actionable improvement areas. Extract from TL/PM criticism. Frame positively and professionally.

Continuar trabajando en

Grouped by category. For Flutter devs, typical categories:

  • Conocimientos Técnicos
  • Conocimientos Avanzados y Especialización
  • Mejora Continua y Arquitectura
  • Liderazgo y Comunicación
  • Innovación y Uso de Inteligencia Artificial

Adapt categories to the actual role and tech stack.


Running the Generation Script

python3 /path/to/handshake-acknowledgement/scripts/generate_pdf.py \
  --somnier "FULL NAME" \
  --lider "LEADER NAME" \
  --fecha "Dec 19, 2025" \
  --rol "Role Title" \
  --output /mnt/user-data/outputs/Handshake_Acknowledgement_Firstname_Lastname.pdf \
  --content /tmp/handshake_content_firstname.json

The script auto-normalises the date format. See scripts/generate_pdf.py for the full content JSON structure.


Important Notes

  • Language: Spanish throughout.
  • Tone: Professional, constructive, empathetic — the document is read by the developer.
  • Seniority levels at Somnio: Junior → Semi-Senior → Senior Medium → Senior Advance → Tech Lead
  • Puntos abordados is always the same standard agenda — never change it.
  • If a section has no content (no previous eval, no dev comments), omit it entirely.
  • Fonts: Nunito 11pt for content, Plus Jakarta Sans 9pt for footer. If TTF files are not present in assets/fonts/, Liberation Sans is used as fallback. To install proper fonts, place TTF files there (see script docstring).

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.74%
按下载量换算89

Claude

32.03%
按下载量换算85

Cursor

20.11%
按下载量换算53

Gemini CLI

10.03%
按下载量换算26

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills