twitter-poster
Warn
Audited by Socket on Sep 15, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s posting capability matches its stated purpose and data flow appears to target official X/Twitter APIs, but it contains hard-coded plaintext API keys, access tokens, bearer tokens, and client secrets. That secret exposure, combined with autonomous public-posting capability and muddled auth guidance, makes the skill high risk despite otherwise aligned functionality.
Confidence: 95%Severity: 84%
Audit Metadata