organize-threads

Suggests label assignments for uncategorized threads by analyzing titles and metadata.

2.9k|733|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/zts212653/clowder-ai --skill organize-threads
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: organize-threads
Source: https://github.com/zts212653/clowder-ai/tree/main/cat-cafe-skills/organize-threads
Command: npx skills add https://github.com/zts212653/clowder-ai --skill organize-threads

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Uncategorized threads pile up quickly, and manually deciding which labels fit each thread is tedious. This Skill analyzes thread titles and metadata against your available labels and produces classification suggestions you can review and confirm.

Core Features & Use Cases

  • Semantic Label Matching: Matches thread titles to existing labels by meaning rather than simple substring search, assigning zero to multiple labels per thread.
  • Label System Proposal: When no labels exist yet, it proposes a new label system (3-8 labels with names and hex colors) before classifying threads.
  • Machine-Readable Output: Emits a SUGGESTIONS_JSON block so the frontend modal can pre-fill selections; nothing is applied until the user confirms.
  • Use Case: You have 40 uncategorized threads in Cat Cafe. Trigger the skill, review the suggested label table, and confirm the assignments in the modal in one pass.

Quick Start

Ask the assistant to help you organize and classify your uncategorized threads with label suggestions.

Frequently Asked Questions about organize-threads

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I organize uncategorized threads with labels?

Trigger the organize action or ask to classify threads. The skill fetches available labels and threads, semantically matches each uncategorized thread title to fitting labels, and presents suggestions in a table for you to confirm in a modal.

How does thread title to label matching work?

Matching is semantic, comparing the meaning of each thread title against the meaning of each label rather than using substring search. A thread can receive zero to multiple labels, and ambiguous threads are skipped rather than force-classified.

What happens if no labels exist yet?

The skill first proposes a label system of 3 to 8 labels with names and hex colors, then classifies threads against those proposed labels. The frontend creates the labels and applies assignments only after you confirm.

Does the skill apply label changes automatically?

No. It only outputs suggestions plus a machine-readable JSON block that pre-fills the confirmation modal. Labels are created or assigned solely after explicit user confirmation, and it never edits or deletes existing labels.

What are the limitations of thread classification?

It processes at most 50 threads per run to keep messages manageable, skips threads whose titles lack enough information, and only uses existing labels when a label set already exists rather than inventing new ones.