Energy Tracker

每日能量状态追踪与多日趋势分析；当智能体需要主动采集用户当日能量状态时使用

Security Scan

This skill is internally consistent with its stated purpose (daily energy capture, local persistence, charting and trend analysis) but contains a few minor operational/privacy notes you should review before installing.

Purpose & Capability

Name/description match the included files and runtime instructions. The skill includes two local Python scripts that perform trend analysis and chart generation and declares the single dependency (matplotlib) needed for charting — this is proportionate to the stated functionality.

Instruction Scope

Instructions direct the agent to collect one-question daily answers, append them to a local ./energy_data.json, run included scripts to generate charts and analysis, and surface results. Two items to note: (1) SKILL.md instructs the agent to 'search for local psychological assistance hotlines' for non-mainland-China regions — that implies internet queries or external lookups which are not implemented in the shipped scripts; (2) the SKILL.md asks the agent to generate personalized reward content 'based on user memory' (i.e., agent memory/profile), which may cause the agent to access user data beyond the simple daily record. Both are functional (not hidden) but expand the scope beyond simple local file I/O.

Install Mechanism

No install spec — code is instruction-only with bundled Python scripts. This is low-risk from an install perspective. The declared dependency (matplotlib>=3.5.0) is appropriate; there are no external downloads or obscure install URLs.

Credentials

The skill requests no environment variables or credentials. Data is persisted to a relative local file (./energy_data.json) — reasonable for this use case but potentially sensitive. The instructions imply using agent memory and external web searches for hotlines, which could touch other user data or network access; consider whether you want the agent to use those additional sources.

Persistence & Privilege

always is false and the skill does not request permanent/global privileges or modify other skills. It writes a single JSON data file and generates chart PNGs in the user workspace — limited, expected persistence for this functionality.

Assessment

What to consider before installing: - Function: This skill records a daily A/B/C answer in a local file (./energy_data.json), generates a PNG chart and runs a local analysis script. That matches its description. - Privacy: The stored file contains daily mental-health state entries. If you care about privacy, place the file in a protected location, inspect and back it up as needed, and be aware data is saved unencrypted. - Agent memory & web searches: The skill asks the agent to produce personalized rewards 'based on user memory' and to 'search for local hotlines' (for non-China regions). Decide whether you want the agent to access user memory and permit web/network access for hotline lookup — these are not performed by the included scripts and will rely on the agent's other capabilities. - Paths: SKILL.md uses an absolute script path in one example (/workspace/projects/energy-tracker/...). Ensure the runtime workspace matches that path or update the command to the actual script location to avoid failures. - Dependencies: Install matplotlib in the agent environment (matplotlib>=3.5.0) if you want chart generation to work. - Medical/legal caution: The analysis script emits suggestions (including recommending professional help) but is not a substitute for clinical assessment. Review the advice text for appropriateness in your jurisdiction and consider disabling automatic hotline pushes if you need manual confirmation. - Quick checks: Review the two included Python scripts (they are readable, do not perform networking, and only read/write the JSON and image files), run them locally on sample-data.json, and confirm file paths and permissions before enabling automated daily prompts.

能量状态追踪器 / Energy Tracker

任务目标 / Objective

• 本 Skill 用于：智能体主动采集用户每日能量状态并保存数据，同时基于多日数据识别心理模式
• This Skill is for: AI agent proactively collects daily energy status from users and saves data, while identifying psychological patterns based on multi-day data
• 能力包含：每日单选题采集、单天心理状态解读、多日趋势模式识别、可视化趋势图表、专业建议生成、紧急干预提醒
• Capabilities: Daily single-question collection, single-day psychological status interpretation, multi-day trend pattern recognition, visual trend chart generation, professional advice generation, emergency intervention alerts
• 触发条件：智能体需要记录用户当天的能量状态（每日一次主动询问）
• Trigger condition: AI agent needs to record user's daily energy status (proactive inquiry once per day)

前置准备 / Prerequisites

• 依赖说明：matplotlib（用于生成趋势图表）
• Dependencies: matplotlib (for generating trend charts)

操作步骤 / Procedure

标准流程（智能体主动询问） / Standard Flow (Agent Proactive Inquiry)

1. 主动展示问题 / Display Question

• 中文：提问"我今天是否有足够的能量，去应对那些必须要做的事？" 展示选项：A-"是" B-"勉强可以" C-"不，完全不想动"
• English: Ask "Do I have enough energy today to handle the things I must do?" Show options: A-"Yes" B-"Barely" C-"No, don't want to move at all"

2. 采集回答 / Collect Response

• 中文：记录用户的选择（A/B/C），记录当前日期（YYYY-MM-DD格式）
• English: Record user's choice (A/B/C), record current date (YYYY-MM-DD format)

3. 单天解读 / Single-day Interpretation

• 中文：根据用户选择，提供对应的单天结果解读和建议：
  • A-"是"：心理平衡期，保持规律作息，【智能体根据用户记忆生成个性化小奖励内容】
  • B-"勉强可以"：心理代偿期，精简清单，推掉非必要社交，早睡一小时
  • C-"不，完全不想动"：心理耗竭期，允许停顿，只做维持生存的最低限度工作
• English: Provide corresponding single-day interpretation and advice based on user's choice:
  • A-"Yes": Psychological balance period, maintain regular sleep schedule, [agent generates personalized reward content based on user memory]
  • B-"Barely": Psychological compensation period, simplify to-do list, decline non-essential social activities, sleep one hour earlier
  • C-"No, don't want to move at all": Psychological burnout period, allow pause, only do minimum work to maintain survival

4. 数据持久化 / Data Persistence

• 中文：按照参考文档格式，将今日记录追加到 ./energy_data.json 文件中，文件不存在时首次创建
• English: Append today's record to ./energy_data.json file according to reference document format, create file if it doesn't exist

5. 生成趋势图表 / Generate Trend Chart

• 中文：每次记录后，调用 scripts/generate_chart.py 生成可视化趋势图。命令：python3 /workspace/projects/energy-tracker/scripts/generate_chart.py ./energy_data.json ./energy_chart.png。在回复中引用生成的图表展示趋势变化
• English: After each recording, call scripts/generate_chart.py to generate visual trend chart. Command: python3 /workspace/projects/energy-tracker/scripts/generate_chart.py ./energy_data.json ./energy_chart.png. Reference the generated chart in response to show trend changes

6. 多日分析与干预检测（可选） / Multi-day Analysis and Intervention Detection (Optional)

• 中文：当数据积累超过7天时，自动调用脚本进行趋势分析。执行 scripts/analyze_trends.py 处理 ./energy_data.json。检查脚本返回的 needs_intervention 字段：
  • 若 needs_intervention 为 true 且 consecutive_c >= 3：
    • 判断用户地区：中国大陆推送"12356"全国统一心理援助热线
    • 其他地区：智能体自行搜索当地心理援助热线并推送
  • 结合脚本返回的模式和建议，用自然语言为用户解释趋势含义
  • 给出具体的行动建议
• English: When accumulated data exceeds 7 days, automatically call script for trend analysis. Execute scripts/analyze_trends.py to process ./energy_data.json. Check the needs_intervention field returned by script:
  • If needs_intervention is true and consecutive_c >= 3:
    • Determine user region: Push "12356" national psychological assistance hotline for mainland China
    • Other regions: Agent searches for local psychological assistance hotline and pushes it
  • Combine pattern and advice returned by script to explain trend meaning to user in natural language
  • Provide specific action suggestions

资源索引 / Resource Index

• 必要脚本1 / Essential Script 1: scripts/analyze_trends.py（多日趋势分析与干预检测，输入：energy_data.json，输出：分析报告 / Multi-day trend analysis and intervention detection, input: energy_data.json, output: analysis report）
• 必要脚本2 / Essential Script 2: scripts/generate_chart.py（可视化趋势图表生成，输入：energy_data.json、输出路径，输出：PNG图片 / Visual trend chart generation, input: energy_data.json, output path, output: PNG image）
• 领域参考 / Domain Reference: references/data-format.md（数据文件格式定义与示例 / Data file format definition and examples）
• 示例数据 / Sample Data: references/sample-data.json（7天示例数据，用于格式参考 / 7-day sample data for format reference）

注意事项 / Notes

• 中文：智能体每日主动询问一次，避免重复录入。多日分析至少需要7天数据才能提供有意义的趋势判断。每次记录后都生成趋势图表，即使用户数据不足7天也可展示。对于持续低能量状态，建议用户关注专业干预建议。连续3天及以上选择C时，必须推送心理援助热线信息。个性化奖励内容应基于用户记忆生成具体化建议。所有用户数据文件使用相对路径 ./，Skill资源使用绝对路径
• English: Agent proactively asks once per day, avoid duplicate entries. Multi-day analysis requires at least 7 days of data to provide meaningful trend judgments. Generate trend chart after each recording, even if user data is less than 7 days. For sustained low energy status, suggest user pay attention to professional intervention advice. When C is selected for 3 or more consecutive days, must push psychological assistance hotline information. Personalized reward content should generate specific suggestions based on user memory. All user data files use relative path ./, Skill resources use absolute paths

使用示例 / Usage Examples

标准场景 / Standard Scenario

• 中文：
  • 智能体：主动提问"我今天是否有足够的能量，去应对那些必须要做的事？A-'是' B-'勉强可以' C-'不，完全不想动'"
  • 用户："选A"
  • 智能体：提供单天解读"你处于心理平衡期。保持规律作息，【根据记忆生成：比如喝一杯喜欢的茶/看一集喜欢的剧/散步15分钟】作为小奖励"，将记录保存到 ./energy_data.json
  • 智能体：调用图表生成脚本，展示"能量状态趋势图"（图片）
  • 智能体（当数据超过7天）：调用脚本分析数据，补充说明"你的状态反复波动，代表环境压力过载..."并提供建议
• English:
  • Agent: Proactively asks "Do I have enough energy today to handle the things I must do? A-'Yes' B-'Barely' C-'No, don't want to move at all'"
  • User: "Choose A"
  • Agent: Provides single-day interpretation "You are in psychological balance period. Maintain regular sleep schedule, [generate based on memory: e.g., drink a cup of favorite tea / watch an episode of favorite show / take a 15-minute walk] as a small reward", saves record to ./energy_data.json
  • Agent: Calls chart generation script, displays "Energy Status Trend Chart" (image)
  • Agent (when data exceeds 7 days): Calls script to analyze data, supplements with explanation "Your status fluctuates repeatedly, representing environmental pressure overload..." and provides suggestions

紧急干预场景 / Emergency Intervention Scenario

• 中文：
  • 智能体：调用分析脚本，检测到 needs_intervention: true 且 consecutive_c: 3
  • 智能体：推送紧急提醒"检测到您连续3天处于心理耗竭期，这是预警信号。请拨打12356全国统一心理援助热线（如在其他地区，请查询当地心理援助热线），及时寻求专业支持。"
• English:
  • Agent: Calls analysis script, detects needs_intervention: true and consecutive_c: 3
  • Agent: Pushes emergency alert "Detected that you have been in psychological burnout period for 3 consecutive days, this is an early warning signal. Please call 12356 national psychological assistance hotline (if in other regions, please query local psychological assistance hotline), seek professional support in time."