← 返回 Skills 市场
142
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install feedback-collector
功能描述
Collects and processes user feedback on task completion. Use after important tasks to ask if the result was satisfactory.
使用说明 (SKILL.md)
Feedback Collector
收集和处理用户对任务完成情况的反馈。
触发时机
重要任务完成后自动触发：
- 文件操作完成
- 内容生成完成
- 配置修改完成
- 技能安装完成
反馈收集方式
简单反馈：
"老板，这个结果满意吗？
A) 满意 ✅
B) 需要改进 ⚠️
C) 不满意 ❌"
详细反馈（当选择 B/C 时）：
"哪里需要改进？
A) 回复太长了
B) 回复太简短了
C) 执行方式不对
D) 结果不符合预期
E) 其他（请说明）"
反馈处理
满意 (A)
- 记录成功模式到
memory/feedback-log.md - 强化当前行为模式
- 回复："好的，记住了！"
需要改进 (B)
- 记录具体问题
- 询问具体改进方案
- 更新用户偏好记录
不满意 (C)
- 记录失败案例到
memory/error-log.md - 提供补救方案
- 道歉并学习
数据存储
# memory/feedback-log.md
## 2026-04-02
### 成功反馈
- 10:30 文件整理任务 ✅ 用户满意
- 14:00 文案生成 ✅ 用户满意
### 改进反馈
- 11:00 回复太长 ⚠️ 已调整为简短模式
- 16:00 执行前未确认 ⚠️ 已添加确认步骤
### 失败反馈
- 09:00 文件路径错误 ❌ 已修复
反馈分析
每周自动生成反馈报告：
- 满意度趋势
- 常见问题类型
- 改进建议
使用示例
任务完成后：
管家：老板，文件整理完成了，把 15 个文件分类到 4 个文件夹。
满意吗？
A) 满意 ✅
B) 需要改进 ⚠️
C) 不满意 ❌
用户：A
管家：好的，记住了！
安全使用建议
This skill appears to do what it claims, but review where the agent stores 'memory/feedback-log.md', 'memory/error-log.md', and 'USER-PREFERENCES.md' on your system and how long those logs are kept. If you handle sensitive content, ensure logs are not synced externally or shared. If you prefer stricter control, require explicit user confirmation before the skill auto-triggers after tasks or limit the task types that prompt feedback. Finally, consider adding a clear retention/privacy policy or a way to purge stored feedback.
功能分析
Type: OpenClaw Skill
Name: feedback-collector
Version: 1.0.0
The feedback-collector skill is a set of markdown instructions designed to guide an AI agent in gathering user satisfaction data after task completion. It defines a workflow for logging feedback into local files like 'memory/feedback-log.md' and adjusting behavior based on user input. There is no executable code, no network activity, and no evidence of malicious intent or data exfiltration.
能力评估
Purpose & Capability
Name/description match the instructions: the SKILL.md describes asking users for simple/detailed feedback, recording logs, and generating reports. Nothing requested (no env vars, no binaries, no installs) is out of scope for a feedback helper.
Instruction Scope
Instructions tell the agent to automatically trigger after important tasks, prompt for A/B/C feedback, and append records to memory/feedback-log.md and memory/error-log.md and update USER-PREFERENCES.md; this is expected for a feedback skill but is somewhat vague about what exactly is written and retention/format. The auto-triggering behavior gives the agent broad discretion to ask after many task types—benign but worth reviewing if you want stricter boundaries.
Install Mechanism
Instruction-only skill with no install spec and no bundled code — lowest-risk pattern. The regex scanner had no files to analyze beyond plain docs.
Credentials
No environment variables, credentials, or external endpoints are requested. The only resources referenced are local memory files (memory/feedback-log.md, memory/error-log.md, USER-PREFERENCES.md), which is proportional to the stated purpose.
Persistence & Privilege
always:false and no install means the skill doesn't demand permanent inclusion. Model invocation is allowed (platform default) which lets the agent invoke the skill autonomously when appropriate—this aligns with the described auto-trigger behavior.
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 feedback-collector - 安装完成后，直接呼叫该 Skill 的名称或使用
/feedback-collector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
feedback-collector 1.0.0 — initial release
- Introduces a feedback collection and processing system for task completion.
- Automatically requests user satisfaction after important tasks.
- Supports both quick and detailed feedback options.
- Logs feedback to memory for continuous learning and improvement.
- Provides weekly feedback analysis and actionable suggestions.
- Chinese and Markdown-based user prompts and response handling included.
元数据
常见问题
Feedback Collector 是什么？
Collects and processes user feedback on task completion. Use after important tasks to ask if the result was satisfactory. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 142 次。
如何安装 Feedback Collector？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install feedback-collector」即可一键安装，无需额外配置。
Feedback Collector 是免费的吗？
是的，Feedback Collector 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Feedback Collector 支持哪些平台？
Feedback Collector 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Feedback Collector？
由 wqz666（@wqz666）开发并维护，当前版本 v1.0.0。
推荐 Skills