Smart Report Generator

Automatically generate and schedule daily, weekly, or monthly reports with customizable templates and multi-platform IM push support.

Key things to consider before installing or running: - Undeclared credential: report_bot.py reads OPENCLAW_API_KEY from the environment but the registry and SKILL.md don't mention this. If you run it and set that key, the script will call the OpenClaw API — only provide that key if you trust the service and the skill author. - Missing dependency documentation: SKILL.md tells you to pip install 'openclaw lark' but the script also needs requests and pyyaml (PyYAML). Add/verify these packages before running. - Feature mismatches: The markdown and description claim scheduling and Slack/企业微信 support, but the code does not implement scheduling and only implements Feishu and Dingtalk. Treat those claims as inaccurate until the author fixes them. - Webhook safety: The config requires a webhook URL. Webhooks are sensitive; do not paste production or privileged webhooks without understanding where messages will be sent and who can receive them. - Audit the code and run in a sandbox: The script is short and readable, but you should inspect the code yourself (or run it in an isolated environment) to confirm there are no hidden endpoints or unexpected network calls. Confirm what OpenClaw account the API key will be billed to and what data is sent to the API (weekly summary prompt includes full task list). - What would change this assessment: if the author updates registry metadata to declare OPENCLAW_API_KEY as a required credential, updates SKILL.md to list all Python dependencies and clearly documents supported platforms and scheduling behavior (or implements these features), then the inconsistencies would be resolved and confidence would increase.

Type: OpenClaw Skill Name: smart-report-generator Version: 1.0.0 The skill bundle is a legitimate utility for generating and distributing work reports. The Python script `report_bot.py` implements the stated functionality by processing task data, optionally using an AI model for summarization, and sending the output to user-configured webhooks (Feishu/DingTalk). No evidence of malicious intent, data exfiltration, or unauthorized execution was found.