meeting-score

会议议题评分系统。会议主持人提供议题名称、材料链接和评委名单，自动在飞书多维表格中创建评分表。每位评委独立打分（打分维度一（替换为实际）、打分维度二（替换为实际）、打分维度三（替换为实际）），打分范围 H/M/L，评分完成后主持人可查看各议题平均分汇总。

Before installing or enabling this skill: - Confirm how the agent will authenticate to Feishu. The SKILL.md uses an {app_token} and API calls but the skill metadata lists no required credentials — ask the platform owner which Feishu token/account will be used and ensure it has least privilege (only access to the specific Bitable app/tables). - Be aware the skill will delete default fields and empty rows when initializing tables; back up any important data or test in a sandbox space first. - The skill creates a cron polling job (recommended every 30s in the instructions). Consider lowering its frequency (e.g., 1–5 minutes) to reduce load and risk, and verify who can view/stop those jobs. - The row-level permission setup requires the meeting host to manually enable Feishu "Advanced permissions" and ensure judge names match Feishu display names (or use member-type fields). Confirm this workflow is acceptable for your org. - Require explicit disclosure of where jobIds and any tokens are stored and who can manage/disable the scheduled tasks. If you cannot get clear answers about which credentials will be used and who controls the scheduled jobs, treat the skill as risky and avoid enabling it system-wide.

Type: OpenClaw Skill Name: meeting-score Version: 1.0.0 The skill is designed for a legitimate meeting scoring system using Feishu Bitable. However, it is classified as 'suspicious' due to its use of powerful capabilities. Specifically, the skill instructs the agent to use the `cron` tool to schedule an `agentTurn` message every 30 seconds. This allows the agent to programmatically instruct itself to perform actions periodically, which, while used for a benign polling mechanism here, represents a significant capability that could be leveraged for malicious purposes if the skill's instructions were altered. Additionally, the skill requires broad permissions to create, read, update, and delete tables, fields, and records within Feishu Bitable, which are high-risk capabilities, even if plausibly needed for its stated purpose.