daily-reflection-journal

每日反思日记引导技能，基于《认知觉醒》的反思框架。当用户说"做每日反思"、"写反思日记"、"今天反思"、"daily reflection"、"回顾今天"等触发。引导用户通过事实-感受-启发三层面深度反思，将日常经历转化为成长素材。

Security Scan

The skill is an instruction-only journaling/reflection guide that only uses local reference files and asks the agent to save markdown entries — its declared requirements and instructions are consistent with its stated purpose.

Purpose & Capability

Name/description (daily reflection journal) match the actual contents: a guided questioning flow plus templates and reference material. The skill requires no binaries, no credentials, and includes only local reference docs, which is proportionate for a journaling helper.

Instruction Scope

Runtime instructions are narrowly scoped to guiding the user through reflection, using the included prompts and theory files. The only non-interactive action specified is saving output to daily-reflection/YYYY-MM-DD.md (a local file). This file write is expected for a journaling skill but is worth noting because it causes the agent to persist user content locally.

Install Mechanism

No install spec and no code files — instruction-only — so nothing is downloaded or executed. This is the lowest-risk install model and appropriate for the skill's purpose.

Credentials

The skill declares no environment variables, credentials, or config paths and its instructions do not reference external secrets or unrelated system configuration. That is proportionate for a personal-reflection skill.

Persistence & Privilege

always is false and the skill does not request persistent or elevated platform privileges. It does instruct the agent to write journal files under a local path, which is normal for this use case and does not modify other skills or global agent settings.

Assessment

This skill appears to be what it says: a local, instruction-only journaling guide. Before installing, confirm where your agent stores files (the skill writes to daily-reflection/YYYY-MM-DD.md) and whether those files are backed up, encrypted, or accessible to others on the system. If your journal entries are sensitive, ensure your agent's runtime environment has appropriate file permissions and storage privacy. There are no network calls, credentials, or installs in the skill, so there is no evidence of external data exfiltration — still verify your agent's broader policies (autonomous invocation, log retention) if you want the agent to save personal content automatically.

每日反思日记

反思 ≠ 流水账。每日反思的核心是捕捉触动你的瞬间，深挖背后的原因，而非记录"今天做了什么"。

反思流程

第一步：选择事件

问用户："今天哪件事让你印象最深、情绪波动最大，或有所触动？"

• 只需 1-3 件事，不求全
• 选择那些有"感觉"的时刻，而非流水账

第二步：三层面反思

对每个事件，依次引导：

1. 事实层面

• 具体发生了什么？（场景、人物、对话）
• 尽量还原当时的情况

2. 感受层面

• 当时的情绪是什么？
• 为什么会有这种感受？
• 至少问自己 3 次"为什么"找到深层原因

3. 启发层面

• 这件事揭示了我怎样的思维模式或行为习惯？
• 如果重来，我会怎么做？
• 我从中学到了什么？
• 以后如何避免或复现？

第三步：提炼行动

• 一个明确的改变：明天开始我会...
• 避免"知道了但不改变"

引导问题

当用户不知如何开始时，使用 prompts-library.md 中的问题启发。

核心原则：

• 随机选 2-3 个问题即可
• 根据回答继续追问"为什么"
• 问题是启发思考，不是评判

理论依据

详见 cognitive-awakening.md

核心要点：

• 元认知：反思是锻炼元认知的最佳方式
• 触动点 = 成长点：抓住情绪波动的瞬间
• 深度学习：将经历转化为经验

输出格式

将反思内容保存到 daily-reflection/YYYY-MM-DD.md，格式：

# YYYY-MM-DD 每日反思

## 事件一：[简短标题]

**事实**：...

**感受**：...

**启发**：...

**行动**：...

## 事件二：[简短标题]

...

## 今日金句

> [一句总结今天的感悟]

## 明日一个小改变

[具体的、可执行的一个改变]

避免的误区

• ❌ 记成流水账（"今天早上起床，吃了早餐..."）
• ❌ 只记录不反思（没有深挖原因）
• ❌ 反思后无行动（知道了但不改变）
• ❌ 求多不求深（一天记录太多事件）