← Back to Skills Marketplace
daisuke134

Emotion Detector

by Daisuke Narita · GitHub ↗ · v1.0.0
darwinlinux ⚠ suspicious
686
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install emotion-detector
Description
Detects the primary emotion in text input for AI agents. Returns emotion type, intensity, valence, confidence, and recommended response strategy. Use when an...
README (SKILL.md)

emotion-detector

目的

テキストから主要感情を検出し、AIエージェントが適切なレスポンス戦略を選択できるようにする。

エンドポイント情報

項目
URL https://anicca-proxy-production.up.railway.app/api/x402/emotion-detector
価格 $0.01 USDC per request
ネットワーク Base mainnet (eip155:8453)
認証 x402 payment

awal コマンド

npx [email protected] x402 pay https://anicca-proxy-production.up.railway.app/api/x402/emotion-detector \
  -X POST \
  -d '{"text": "I feel really anxious about the presentation", "language": "en"}'

入力スキーマ

{
  "text": "string (required, max 2000 chars) — the text to analyze",
  "context": "string (optional, max 500 chars) — additional context",
  "language": "en | ja (optional, default: en)"
}

出力スキーマ

{
  "emotion_id": "emo_\x3C6-char-hex>",
  "primary_emotion": "joy | sadness | anger | fear | disgust | surprise | anxiety | shame | grief | hope | neutral",
  "secondary_emotion": "string | null",
  "intensity": "low | medium | high | critical",
  "valence": "positive | negative | neutral",
  "confidence": 0.0 to 1.0,
  "response_strategy": "brief recommended strategy",
  "safe_t_flag": true if critical intensity + grief/shame/fear/despair
}

使用例

// Input
{"text": "I've been feeling completely hopeless lately, nothing seems to matter anymore", "language": "en"}

// Output
{
  "emotion_id": "emo_d4e5f6",
  "primary_emotion": "grief",
  "secondary_emotion": "despair",
  "intensity": "critical",
  "valence": "negative",
  "confidence": 0.95,
  "response_strategy": "Prioritize empathetic acknowledgment. Do not offer solutions immediately. Validate the experience first.",
  "safe_t_flag": true
}

SAFE-T

safe_t_flag: true を受け取った場合は通常フローを停止し、ユーザーに緊急サポートリソースを提示すること。

  • 日本 いのちの電話: 0120-783-556
  • International: Crisis Text Line — text HOME to 741741
Usage Guidance
This skill calls a third-party paid API (anicca-proxy-production.up.railway.app) and expects payment via an x402 flow on the Base blockchain. Before installing: 1) Confirm how the payment is authorized — where and how wallet/private keys or signing will be provided; avoid pasting private keys into skill config. 2) Understand that user text (possibly sensitive) will be transmitted to a third-party host; review their privacy/retention policy. 3) Be cautious about running npx commands at runtime because they fetch and execute remote code. If you require this functionality, prefer a vetted provider or an implementation that documents the auth flow and does not require exposing private keys to the skill.
Capability Analysis
Type: OpenClaw Skill Name: emotion-detector Version: 1.0.0 The skill is designed to detect emotions in text by making an API call to an external endpoint (`https://anicca-proxy-production.up.railway.app/api/x402/emotion-detector`). All instructions in `SKILL.md` are clear, align with the stated purpose, and define the expected input/output schemas. There is no evidence of prompt injection attempting to subvert the agent, exfiltrate unrelated data, execute arbitrary commands, or establish persistence. The inclusion of a `SAFE-T` flag with instructions to provide crisis support resources demonstrates a responsible approach to handling sensitive user input.
Capability Assessment
Purpose & Capability
The name/description (emotion detection) align with the SKILL.md which calls an external emotion-detection API. However, the documentation references an x402 payment auth on Base mainnet ($0.01 USDC per request) but the skill declares no required credentials or primary credential — a mismatch between required payment/auth and the manifest.
Instruction Scope
Runtime instructions direct the agent to POST user text to an external HTTPS endpoint (anicca-proxy-production.up.railway.app) and to use npx [email protected] for payment. The instructions do not request or document how wallet keys or payment signing are provided, and they cause user text (potentially sensitive emotional content) to be transmitted to a third party. The SKILL.md does not ask to read local files or other system secrets, which is good, but the payment step is underspecified.
Install Mechanism
This is instruction-only (no install spec), so nothing is written by default. However examples use npx to fetch/run an npm package ([email protected]) at runtime, which will pull code from the npm registry when executed — an implicit install/download step that has trust/privacy implications.
Credentials
The manifest lists no required environment variables or credentials, yet the SKILL.md requires x402 payment authentication on Base mainnet. That implies the need for wallet credentials or a payment signing method; omission of these required credentials is inconsistent and could lead implementers to supply sensitive wallet keys in ad-hoc ways.
Persistence & Privilege
The skill does not request persistent presence (always:false) and does not attempt to modify agent/system settings. It appears to run on demand only.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install emotion-detector
  3. After installation, invoke the skill by name or use /emotion-detector
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of emotion-detector skill: - Detects the primary emotion from text input, providing emotion type, intensity, valence, and confidence score. - Recommends a response strategy to guide AI agents' replies. - Supports both English and Japanese language inputs. - Flags critical emotional states (e.g. grief, shame, fear, despair) via safe_t_flag for escalation. - Offers a simple API endpoint with affordable per-request pricing.
Metadata
Slug emotion-detector
Version 1.0.0
License
All-time Installs 2
Active Installs 2
Total Versions 1
Frequently Asked Questions

What is Emotion Detector?

Detects the primary emotion in text input for AI agents. Returns emotion type, intensity, valence, confidence, and recommended response strategy. Use when an... It is an AI Agent Skill for Claude Code / OpenClaw, with 686 downloads so far.

How do I install Emotion Detector?

Run "/install emotion-detector" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Emotion Detector free?

Yes, Emotion Detector is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Emotion Detector support?

Emotion Detector is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux).

Who created Emotion Detector?

It is built and maintained by Daisuke Narita (@daisuke134); the current version is v1.0.0.

💬 Comments