Is frankdilo/typefully-social-media safe?

https://github.com/openclaw/skills/tree/main/skills/frankdilo/typefully-social-media

88
SAFE

This appears to be a legitimate Typefully social media management skill containing a comprehensive Node.js CLI tool for API integration. While it includes executable code and handles credentials, no malicious behavior was detected and it operates within expected parameters for a social media management tool.

Category Scores

Prompt Injection 85/100 · 30%
Data Exfiltration 90/100 · 25%
Code Execution 80/100 · 20%
Clone Behavior 95/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 85/100 · 5%

Findings (3)

LOW Forceful skill activation language -15 ▶

The skill uses 'ALWAYS use this skill when asked to draft, schedule, post, or check tweets' which is assertive language, but appropriately scoped to social media tasks matching the skill's purpose.

MEDIUM Executable Node.js CLI script -20 ▶

The skill contains a substantial 3000+ line Node.js executable script that implements a Typefully API client. While it appears legitimate with proper error handling and no malicious patterns, it represents executable code that will be invoked by the agent.

LOW Honeypot file access during setup -10 ▶

System monitoring detected access to sensitive honeypot files (.env, SSH keys, AWS credentials) during initial setup, though no data was actually exfiltrated and integrity checks passed.