# Sentinel Mood

> Analyze the sentiment and emotional tone of text using NLTK and VADER. Use this to gauge user mood, detect urgency, or analyze content tone.

- Skill: `modbender/sentinel-mood` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add modbender/sentinel-mood`
- Raw SKILL.md: https://api.skillmd.com/api/skills/modbender/sentinel-mood/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: modbender (https://skillmd.com/u/modbender)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/modbender/sentinel-mood

---


# Sentinel Mood

A lightweight sentiment analysis skill powered by NLTK's VADER (Valence Aware Dictionary and sEntiment Reasoner). It is specifically tuned for social media texts, conversational language, and short updates.

## Capabilities

- **Analyze Sentiment:** Get positive, negative, neutral, and compound scores for any text.
- **Detect Tone:** (Implicit) Infer tone based on polarity scores.

## Usage

**User:** "Analyze the sentiment of this message: 'I love how this project is turning out, great job!'"
**Agent:** [Runs skill] -> Returns sentiment scores (e.g., compound: 0.8, pos: 0.6).

## Technical Details

This skill uses a Python script (`analyze.py`) that imports `nltk.sentiment.SentimentIntensityAnalyzer`.

### Dependencies

- Python 3
- `nltk` library (`pip install nltk`)
- `vader_lexicon` (downloaded via `nltk.downloader`)

## Implementation

The skill executes a python script that takes text as an argument and outputs JSON.

