Skip to content
Back to skills

Mode Classification

ASecurity

Attribute catalog capability triggers between task prompt and user message in CLI mode raising

  • 10 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agents

Works with

  • cli

Security analysis

A100/100

Scanned November 11, 2026

npx -y skills add D-Shey/autoharness --skill mode-classification --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mode Classification?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Mode Classification
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/d-shey-mode-classification/badge)](https://www.skillsdirectory.com/skills/d-shey-mode-classification)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: mode-classification
description: Attribute catalog capability triggers between task prompt and user message in CLI mode raising
---

# Mode classification and word attribution

## When to use

Use when explaining why a task mode was raised from light to full, identifying trigger keywords, or attributing matched capability terms to either the task prompt or the user's conversation prompt.

## Steps

1. Classify the combined task prompt and user message together (`catalog.classify(about)`).
2. For each capability and its matched trigger words, check where each word appears:
   - Check if the word is in the task text (case-insensitive substring or token match).
   - Check if the word is in the user message (case-insensitive substring or token match).
3. If the user message is empty or the word does not appear in the user message, do not attribute it to "the user's message".
4. Group words by their actual source: "the task text" or "the user's message". If a word appears in both, default to "the task text".
5. Format the diagnostic explanation clearly with capability names and source tags.

## How to verify

Run unit tests verifying that keywords in task text are credited to "the task text", keywords in user message are credited to "the user's message", and absent user message never receives attributed words.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.