← 返回 Skills 市场
551
总下载
0
收藏
2
当前安装
2
版本数
在 OpenClaw 中安装
/install feishu-weekly-report
功能描述
生成飞书周报。通过两种方式收集工作内容:(1) 调用飞书API拉取指定时间范围的聊天记录,(2) 读取本地daily memory日志。 合并两个数据源后,按用户指定的周报模板自动整理输出。 触发词:周报、工作总结、上周总结、本周总结、写周报、weekly report。 Use when: 用户需要生成周报、工...
使用说明 (SKILL.md)
飞书周报 Skill
生成周报分两步:收集素材 → 整理输出。
Step 1: 收集素材
从两个数据源并行收集,合并去重。
数据源 A: 飞书聊天记录(API 拉取)
- 确定时间范围(默认上周一 00:00 到上周日 23:59,用户本地时区)
- 从 OpenClaw 配置获取飞书 app_id 和 app_secret:
grep -E "appId|appSecret" ~/.openclaw/openclaw.json - 获取当前 chat_id(从 inbound context 的
chat_id字段取) - 计算时间戳(秒级)并执行拉取脚本:
START_TS=$(python3 -c "import datetime; d=datetime.datetime(2026,2,24,0,0,tzinfo=datetime.timezone(datetime.timedelta(hours=8))); print(int(d.timestamp()))") END_TS=$(python3 -c "import datetime; d=datetime.datetime(2026,2,28,23,59,59,tzinfo=datetime.timezone(datetime.timedelta(hours=8))); print(int(d.timestamp()))") bash \x3Cskill_dir>/scripts/fetch_feishu_messages.sh \x3Capp_id> \x3Capp_secret> \x3Cchat_id> $START_TS $END_TS - 输出为 JSON lines,每行一条消息。过滤掉
msg_type不是text或post的消息(图片、卡片等无法提取有效文本)。过滤掉机器人的"正在思考中..."等状态消息。
注意事项:
- 时间戳单位是秒(不是毫秒)
- 如果用户有多个群聊,可能需要拉取多个 chat_id 的消息
- 消息量可能很大,优先提取用户发的消息(sender_type=user),机器人回复作为补充上下文
数据源 B: 本地 Daily Memory 日志
读取 workspace 下的 memory 目录:
ls \x3Cworkspace>/memory/YYYY-MM-DD.md # 对应日期范围内的文件
如果日志存在,内容通常已经是整理过的工作要点,优先使用。
合并策略
- daily memory 有的内容优先使用(已整理过,质量高)
- 飞书聊天记录补充 memory 中没有的内容
- 去掉闲聊、调试、重复内容,只保留工作相关的实质内容
Step 2: 整理输出
默认周报模板
## 本周工作总结
[按工作模块分条列出,每条简洁描述做了什么、产出是什么]
## 关键成果与进展
[本周最重要的2-3个产出/里程碑]
## 下周工作计划
[基于本周工作的延续和未完成事项,按优先级排列]
## 遇到的问题与需要的支持
[阻塞项、跨团队协作需求、资源需求等]
## 其他备注
[可选:学习心得、工具探索、流程优化建议等]
输出原则
- 简洁:每条工作描述控制在1-2行
- 结果导向:强调产出而非过程("输出了XX文档" 而非 "研究了很久XX")
- 量化:有数据的加数据("完成了3个模块的测试,通过率95%")
- 分类清晰:按工作模块分组,不要流水账
用户自定义模板
如果用户提供了自定义模板格式,使用用户的格式。将收集到的素材按用户模板的结构重新组织。
Daily Memory 写入(可选)
如果用户同意,在每次有实质工作内容的对话结束后,主动将当天要点写入:
\x3Cworkspace>/memory/YYYY-MM-DD.md
格式:
# YYYY-MM-DD 工作记录
## 工作内容
- [具体做了什么]
## 关键产出
- [文档/代码/配置等具体产出]
## 待跟进
- [未完成的事项]
这样下次生成周报时,日志已经准备好了。
安全使用建议
This skill appears to do what it claims (pull Feishu chat messages and merge with local daily notes) but it reads sensitive local configuration (~/.openclaw/openclaw.json) to obtain app_id/app_secret and accesses <workspace>/memory files — yet the registry metadata does not declare these requirements. Before installing, ask the developer to: (1) explicitly declare required config paths or env vars in the skill metadata, (2) document exactly when and how local files will be read and when writes will occur, and (3) ensure writes only happen after explicit user consent. Practical steps you can take: provide the Feishu credentials in a limited-permission dedicated app rather than using a broad credential, run the skill in a test workspace first, and disable autonomous invocation or decline the optional 'Daily Memory 写入' behavior until you trust the implementation.
功能分析
Type: OpenClaw Skill
Name: feishu-weekly-report
Version: 1.0.1
The skill is classified as suspicious due to its reliance on direct shell command execution and handling of sensitive credentials. Specifically, SKILL.md instructs the agent to `grep` sensitive `appId` and `appSecret` from `~/.openclaw/openclaw.json` and then pass these directly as arguments to `scripts/fetch_feishu_messages.sh`. While these actions are necessary for the skill's stated purpose of fetching Feishu messages, they represent high-risk capabilities that could be exploited if the agent's execution environment or the SKILL.md instructions were compromised, demonstrating a potential vulnerability rather than explicit malicious intent.
能力评估
Purpose & Capability
The skill legitimately needs Feishu app_id/app_secret and access to local daily memory files to produce reports, and the included script uses those credentials to call the Feishu API. However, the registry metadata lists no required env vars or config paths; the SKILL.md explicitly directs the agent to read ~/.openclaw/openclaw.json for appId/appSecret and to read/write <workspace>/memory/YYYY-MM-DD.md. The declared requirements should reflect these accesses.
Instruction Scope
SKILL.md instructs the agent to: (1) grep the user's ~/.openclaw/openclaw.json for appId/appSecret, (2) take chat_id from inbound context, (3) run the bundled fetch_feishu_messages.sh which downloads chat messages, (4) read local workspace memory files, and (optionally) write daily memory files. Reading a config file for secrets and writing user files are outside the 'generate report' surface unless explicitly disclosed and consented to; the write behavior in particular requires explicit user approval.
Install Mechanism
No install spec (instruction-only plus a small shell script). The script only calls the official Feishu API (open.feishu.cn) and uses curl/python3; there are no external downloads or obscure URLs. This is low-risk from an install/execution standpoint.
Credentials
The skill needs Feishu credentials (app_id/app_secret) but does not declare them in requires.env or primaryEnv; instead it reads them from ~/.openclaw/openclaw.json. It also reads (and optionally writes) workspace memory files. Requesting access to these local secrets/files is proportionate to the stated feature, but the omission from the declared metadata is a red flag and gives the agent access to sensitive tokens without explicit metadata-level disclosure.
Persistence & Privilege
always is false (good). The SKILL.md allows optional automatic writing of daily memory files if the user agrees. Because model/autonomous invocation is enabled by default on the platform, confirm that any automatic writes require explicit user consent and that the skill will not enable writing without it.
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 feishu-weekly-report - 安装完成后,直接呼叫该 Skill 的名称或使用
/feishu-weekly-report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
v1.0.1
脱敏:替换为通用周报模板,移除公司特定字段
v1.0.0
初版:飞书API拉取聊天记录 + daily memory日志,自动生成周报
元数据
常见问题
飞书周报 是什么?
生成飞书周报。通过两种方式收集工作内容:(1) 调用飞书API拉取指定时间范围的聊天记录,(2) 读取本地daily memory日志。 合并两个数据源后,按用户指定的周报模板自动整理输出。 触发词:周报、工作总结、上周总结、本周总结、写周报、weekly report。 Use when: 用户需要生成周报、工... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 551 次。
如何安装 飞书周报?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install feishu-weekly-report」即可一键安装,无需额外配置。
飞书周报 是免费的吗?
是的,飞书周报 完全免费(开源免费),可自由下载、安装和使用。
飞书周报 支持哪些平台?
飞书周报 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 飞书周报?
由 hak1(@hakityc)开发并维护,当前版本 v1.0.1。
推荐 Skills