iGPT email ask

• Analyzed: May 12, 2026, 4:22 AM
• Findings: 1
• Version: 1.0.0
• Hash: 0b6d515a2455c605e758471f5ec330acabd9c4dbb346f08887beb2811b5ecb16
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill bundle is benign. The `SKILL.md` clearly outlines the skill's purpose to analyze user email data via the `igpt.ai` API, explicitly disclaiming shell/filesystem access, persistence, and unauthorized data exfiltration. It details a robust security model including API-key scoping, per-user isolation, read-only OAuth, and no data retention. There are no prompt injection attempts against the agent, no signs of malicious execution, or obfuscation. All communication is restricted to `api.igpt.ai`.