← 返回 Skills 市场

Daily Report Skill

Name: Daily Report Skill
Author: rare

作者 raressen · GitHub ↗ · v1.0.0

cross-platform ⚠ suspicious

525

总下载

当前安装

版本数

在 OpenClaw 中安装

/install daily-report-skill

功能描述

Daily report skill - Generate and send daily work summary. Keywords: daily report, work summary, end of day. Use when: End of day reporting, or manually trig...

使用说明 (SKILL.md)

Daily Report Skill

Trigger

Configure via OpenClaw cron to auto-execute at your preferred time
Or manual trigger: "generate daily report"

Workflow

Load config
- Read config/daily-report.json (workspace root)
- Fall back to defaults or prompt user if missing
- See references/config.example.md
Collect data
- Read memory/YYYY-MM-DD.md (today's journal)
- Read conversation logs, task status (optional)
Generate report
- Load template: assets/template.md
- Replace variables: {{date}}, {{time}}, {{recipient}}, {{agent_name}}
- Fill sections: work, learning, review, suggestions, evolution
Save file
- Write to memory/daily-reports/YYYY-MM-DD.md
Send notification
- Iterate through configured channels
- Call corresponding message API per channel type
- Supports parallel multi-channel delivery

Configuration

Config file: config/daily-report.json (workspace root, not included in skill package)

Required fields:

recipient.name - Report recipient name
channels - Notification channels (at least one)
agent_name - Agent name

See references/config.example.md for details.

Template

Report template at assets/template.md

Supported variables:

{{date}} - Date (YYYY-MM-DD)
{{time}} - Time (HH:MM)
{{recipient}} - Recipient name
{{agent_name}} - Agent name

Manual Invocation

generate daily report

Or with temporary overrides:

generate daily report and send to feishu ou_xxx

安全使用建议

Before installing, clarify how messaging credentials are supplied and stored (webhook URL, bot token, or platform-specific credentials) and avoid putting secrets in plain workspace files. Confirm exactly which conversation logs/files the agent will read and restrict that to only what's necessary for the report. If the skill will send messages automatically (cron), test it in a sandboxed workspace or with a test recipient to ensure no sensitive data is leaked. Ask the publisher for explicit documentation of required auth fields and recommended secure storage (e.g., secret manager or platform-provided credentials) — the absence of declared credentials in the package is the main gap here.

功能分析

Type: OpenClaw Skill Name: daily-report-skill Version: 1.0.0 The skill bundle describes a 'daily-report' functionality, including reading agent memory/conversation logs, generating a report from a template, and sending it via configured communication channels (Feishu, Discord, Slack, Telegram). While this involves handling potentially sensitive data and network communication, these actions are explicitly aligned with the stated purpose of generating and sending a daily report. There is no evidence of intentional harmful behavior, such as unauthorized data exfiltration, malicious code execution, persistence mechanisms, or prompt injection attempts in SKILL.md designed to subvert the agent's operation or user's intent. The configuration for recipients is external to the skill bundle and user-controlled.

能力评估

⚠ Purpose & Capability

The skill claims to generate and send daily reports which legitimately requires reading local notes and calling messaging APIs. However, the package declares no environment credentials or config requirements for authenticating to Feishu/Discord/Slack/Telegram; the provided config example only shows recipient IDs, not API keys or webhook URLs. That mismatch (capability to send messages vs. no declared way to authenticate) is unexplained.

⚠ Instruction Scope

Runtime instructions tell the agent to read memory/YYYY-MM-DD.md and 'conversation logs, task status (optional)'. Reading conversation logs can expose broad sensitive context; the SKILL.md gives no limits on which logs or how much context to include. Instructions to 'call corresponding message API per channel type' are vague about endpoints and authentication, leaving implementation and data handling decisions unspecified and potentially risky.

✓ Install Mechanism

This is an instruction-only skill with no install spec and no code files, so it does not write or execute new artifacts on disk on install — low install-surface risk.

⚠ Credentials

No required environment variables or primary credential are declared, yet the skill expects to send messages through external services which normally require tokens, webhooks, or bot credentials. The example config does not include credentials or webhook URLs. This under-specification could lead to credentials being added to workspace config in an ad-hoc, insecure way or the agent being configured to use globally available tokens unexpectedly.

✓ Persistence & Privilege

The skill is not always-enabled and does not request special platform privileges. It writes reports into workspace memory paths under its own scope (memory/daily-reports/) and does not claim to modify other skill configs.

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 daily-report-skill
安装完成后，直接呼叫该 Skill 的名称或使用 /daily-report-skill 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

Configurable daily report with multi-channel delivery

元数据

Slug daily-report-skill

版本 1.0.0

许可证 —

累计安装 4

当前安装数 4

历史版本数 1

常见问题