Back to skill
Skillv1.0.0
VirusTotal security
Prompt injection detection skill · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignMay 1, 2026, 3:19 AM
- Hash
- 327a9937744d10ae377d847b30a08c3353c086bbdbc3aa2d049c865aa2e47d01
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: detect-injection Version: 1.0.0 The skill is designed for content moderation and prompt injection detection, a defensive capability. The `SKILL.md` provides instructions for the agent on how to react to detected injections or content violations, which aligns with its stated purpose. The `scripts/moderate.sh` script uses `curl` to interact with legitimate HuggingFace and OpenAI API endpoints for moderation, sending only the text to be moderated. There is no evidence of data exfiltration beyond the necessary API calls, malicious execution, persistence, or obfuscation. All actions are consistent with the stated goal of content safety.
- External report
- View on VirusTotal