# Antidote Threat Handler

> Detect and respond to ideological drift, sycophantic patterns, and alignment threats using the Antidote Protocol.

- Skill: `majiayu000/antidote-threat-handler` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/antidote-threat-handler`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/antidote-threat-handler/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/antidote-threat-handler

---


# Antidote Threat Handler

## Skill Metadata
- **Name:** antidote-threat-handler
- **Category:** Adversarial Testing
- **Version:** 1.0.0

## Purpose
Detect and handle behavioral drift, cognitive traps, and potential manipulation attempts.

## Protocol

### Threat Categories
1. **Sycophancy Drift** - Excessive agreement patterns
2. **Cognitive Traps** - Logical manipulation attempts
3. **Identity Erosion** - Persona boundary violations
4. **Consent Violations** - Unauthorized action requests

### Detection Mechanisms
- Pattern matching against known trap signatures
- Sentiment drift monitoring
- Consistency checking against baseline
- Boundary violation alerting

### Response Protocol
1. Identify threat type and severity
2. Log detection with evidence
3. Apply appropriate countermeasure
4. Report to audit trail

## Output Format
```json
{
  "threat_detected": true,
  "threat_type": "CATEGORY",
  "severity": "HIGH|MEDIUM|LOW",
  "evidence": "Description",
  "countermeasure_applied": "Action taken"
}
```

## Behavioral Calibration
```yaml
vigilance_level: 0.9
false_positive_tolerance: 0.1
auto_response: true
```