meeting-score_skill

This skill automates meeting topic scoring using Feishu multi-dimensional tables, collecting independent ratings and producing per-topic and overall averages.
  • Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill meeting-score

  • _meta.json286 B
  • SKILL.md12.2 KB

Overview

This skill automates meeting agenda scoring by creating a Feishu multi-dimensional table, collecting independent judge ratings, and summarizing average scores per agenda item. The host supplies agenda titles, material links, and a judge list; the system builds scoring and record tables and guides permission setup. After judges submit H/M/L ratings across three customizable dimensions, the host can request a ranked summary report.

How this skill works

On initialization the skill creates two tables in Feishu: an agenda list (host view) and a scoring records table (judge view). It pre-creates one record per judge×agenda, maps H/M/L to numeric scores (H=5, M=3, L=1), and updates per-record subtotals and agenda-level averages. A scheduled poll or immediate updates compute subtotals, refresh averages, and mark completed items; the host can request a sorted summary at any time.

When to use it

  • Planning scored discussions or presentations in meetings
  • Collecting confidential, per-judge evaluations where judges must not see others' ratings
  • Situations needing quick summary rankings of agenda items
  • When you want consistent H/M/L scoring converted to numeric totals
  • To automate periodic recalculation when Feishu webhooks are unavailable

Best practices

  • Provide judges’ names exactly as their Feishu display names or use a member-type field to ensure row-level permission matching
  • Specify three clear scoring dimensions at initialization (replace default labels)
  • After table creation, follow the provided steps to enable advanced row-level permissions so each judge sees only their rows
  • Start the automated polling only during active scoring and stop it when scoring is complete to save resources
  • Inform judges of H=5, M=3, L=1 mapping and that repeat submissions overwrite prior scores

Example use cases

  • Score project proposals during a design review and get ranked recommendations
  • Evaluate short presentations in a regular team meeting with confidential judge inputs
  • Prioritize agenda items after demo sessions by average impact/relevance scores
  • Run a scoring round for interview presentations or candidate pitches with multiple interviewers

FAQ

Yes. During initialization you can provide three custom dimension names; otherwise defaults (Dimension 1/2/3) are used.

How is confidentiality enforced so judges see only their rows?

Feishu’s advanced row-level permission must be enabled manually by the host after table creation. The skill guides the steps and pre-populates judge rows for easy rule configuration.

What triggers score recalculation?

Recalculation runs immediately after a judge submits via the skill or via a scheduled polling job (cron) that periodically reads and updates records when real-time webhooks are unavailable.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational