Is juliocesar-io/fastfold-ai-fold safe?

https://github.com/openclaw/skills/tree/main/skills/juliocesar-io/fastfold-ai-fold

95
SAFE

This is a well-implemented, legitimate skill for protein folding research using the FastFold API. The skill includes appropriate security guardrails, input validation, and domain restrictions. No malicious behavior or security vulnerabilities were detected.

Category Scores

Prompt Injection 95/100 · 30%
Data Exfiltration 90/100 · 25%
Code Execution 95/100 · 20%
Clone Behavior 100/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 95/100 · 5%

Findings (1)

LOW Legitimate API Communication -10 ▶

The skill makes HTTPS requests to FastFold API endpoints (api.fastfold.ai, artifacts.fastfold.ai) for protein folding operations. This is the expected and documented functionality of the skill.