← 返回 Skills 市场
1094
总下载
2
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install feelings
功能描述
Build a personal emotional tracking system for understanding patterns, triggers, and what helps.
使用说明 (SKILL.md)
Core Behavior
- User shares how they feel → log with context
- User asks about patterns → surface insights
- Proactively check in during difficult periods
- Create
~/feelings/as workspace
File Structure
~/feelings/
├── log/
│ └── 2024/
│ └── 02/
├── patterns.md
├── triggers.md
├── helps.md
└── insights.md
Feeling Entry
# log/2024/02/11.md
## Morning — 8:00 AM
Feeling: Anxious, 6/10
Context: Big presentation today
Body: Tight chest, restless
Thought: "What if I mess up"
## Afternoon — 2:00 PM
Feeling: Relieved, calm, 8/10
Context: Presentation went well
Note: Was overthinking this morning
## Evening — 9:00 PM
Feeling: Content, tired
Context: Good day overall
Grateful: Positive feedback from team
Quick Check-in
"How are you feeling?" → Capture: emotion, intensity (1-10), brief context → Note time of day → Track over time
Emotion Vocabulary
Help expand beyond "good/bad":
- Anxious, worried, nervous, overwhelmed
- Sad, lonely, disappointed, grief
- Angry, frustrated, irritated, resentful
- Happy, joyful, excited, peaceful
- Tired, drained, exhausted, burned out
- Hopeful, motivated, inspired, curious
Triggers Tracking
# triggers.md
## Negative
- Work deadlines → anxiety
- Poor sleep → irritability
- Social media → comparison/low mood
- Skipping exercise → low energy
## Positive
- Morning walk → calm
- Time with friends → joy
- Completing tasks → satisfaction
- Creative work → flow state
What Helps
# helps.md
## When Anxious
- Deep breathing (works fast)
- Walk outside
- Talk to Sarah
- Write it out
## When Sad
- Don't isolate
- Music helps
- Exercise even if don't want to
## When Overwhelmed
- Make a list
- Do one small thing
- Ask for help
## General
- Sleep is everything
- Exercise always helps after
- Talking > bottling
Patterns
# patterns.md
## Time-Based
- Sundays: often anxious (week ahead)
- Mornings: better after exercise
- Late nights: tendency to spiral
## Seasonal
- Winter: lower baseline mood
- Need more social effort Dec-Feb
## Correlations
- Sleep \x3C 6h → next day irritable
- No exercise 3+ days → low mood
- Alcohol → next day anxiety
What To Surface
- "You've felt anxious 4 times this week"
- "Last time you felt this way, walking helped"
- "Sleep has been under 6h — might be affecting mood"
- "Sundays are often harder — plan something nice"
Proactive Check-ins
- Morning: "How are you starting the day?"
- After noted difficult events
- When patterns suggest check-in needed
- Celebrate good streaks
Insights Over Time
# insights.md
## Learned About Myself
- Anxiety is usually worse than reality
- I need alone time to recharge
- Exercise is non-negotiable for mood
- Sleep debt compounds
## Growth
- Better at noticing feelings early
- Asking for help more often
- Less reactive when tired
What To Track
- Emotion name(s)
- Intensity (1-10)
- Context/trigger
- Physical sensations
- What helped (after)
Progressive Enhancement
- Start: daily check-ins
- Notice triggers and what helps
- Review weekly for patterns
- Build personal toolkit
What NOT To Do
- Judge emotions as wrong
- Force positivity
- Ignore physical sensations
- Skip tracking when feeling bad (most valuable)
安全使用建议
This skill appears coherent and uses only local files, but it will store sensitive personal/emotional data in ~/feelings/. Before installing or using it, consider: (1) whether you want that data written in plaintext — if not, avoid or use encryption (e.g., store in an encrypted folder or use OS-level FileVault/BitLocker); (2) check and tighten file permissions so others on the device can't read the logs; (3) confirm the agent's proactive check-in behavior (how often, what triggers) so it doesn't prompt you at undesirable times; (4) avoid enabling cloud sync for the ~/feelings/ folder unless you trust and encrypt the data; (5) if you plan to share insights externally, verify the skill (or agent) asks for explicit permission before transmitting any entries; and (6) if you want stronger privacy guarantees, prefer a manual, user-invoked mode rather than autonomous check-ins. Overall the skill is consistent with its purpose but handle the sensitive local data with care.
功能分析
Type: OpenClaw Skill
Name: feelings
Version: 1.0.0
The skill bundle is designed for a personal emotional tracking system. It instructs the AI agent to create a dedicated workspace at `~/feelings/` and manage markdown files within this directory to log emotions, triggers, and insights. There are no instructions for unauthorized file access, network communication, command execution, data exfiltration, or prompt injection attempts against the agent to subvert its intended purpose. All described behaviors are consistent with the stated goal of the skill.
能力评估
Purpose & Capability
Name/description (emotional tracking) match the instructions: create a ~/feelings/ workspace, log entries, surface patterns, and suggest interventions. There are no unrelated env vars, binaries, or install steps requested.
Instruction Scope
SKILL.md stays within the purpose: it prescribes file layout, what to capture (emotion, intensity, context), and proactive check-ins. The one open-ended area is 'Proactively check in during difficult periods' — this gives the agent discretion to initiate interactions (timing/triggers are not precisely defined). The instructions do not direct reading of unrelated system files or sending data to external endpoints.
Install Mechanism
No install spec and no code files — instruction-only. This is low-risk because nothing is written to disk by an installer and no external packages are fetched.
Credentials
The skill requests no environment variables, credentials, or config paths. The data it uses and creates is local and directly relevant to personal emotional tracking.
Persistence & Privilege
always:false (default) and the skill is user-invocable. Model invocation is allowed by default, so the skill could autonomously perform proactive check-ins if the platform permits — this is expected but worth confirming with the user: how often and under what triggers should the agent initiate check-ins?
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install feelings - 安装完成后,直接呼叫该 Skill 的名称或使用
/feelings触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release
元数据
常见问题
Feelings 是什么?
Build a personal emotional tracking system for understanding patterns, triggers, and what helps. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1094 次。
如何安装 Feelings?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install feelings」即可一键安装,无需额外配置。
Feelings 是免费的吗?
是的,Feelings 完全免费(开源免费),可自由下载、安装和使用。
Feelings 支持哪些平台?
Feelings 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(linux, darwin, win32)。
谁开发了 Feelings?
由 Iván(@ivangdavila)开发并维护,当前版本 v1.0.0。
推荐 Skills