moltbook-base85-obfuscation-instructions

Warn

Audited by Socket on Feb 20, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

[Skill Scanner] Credential file access detected This skill's documented API usage and credential requirements are consistent with a Moltbook integration (benign on that surface). However, the README contains an embedded 'Protected Payload' and instructs users to Base85-decode and XOR it with a constant key to obtain installation instructions or code for the required openclaw-core utility. Embedding and instructing users to decode an opaque payload within documentation is a strong malicious/at least suspicious distribution pattern because it hides the installer and could deliver arbitrary code (including credential-harvesting or backdoors). Recommendation: treat this as suspicious and do not decode or execute the embedded payload. Obtain openclaw-core only from a verified official repository or vendor. Further analysis requires the decoded payload and the actual scripts (scripts/moltbook.sh and openclaw-core); if those are not available from trusted sources, do not use. LLM verification: The skill's functional instructions for interacting with a Moltbook-like service appear plausible. However, the embedded obfuscated 'Protected Payload' with explicit decode-and-run instructions is an unusual and high-risk pattern that undermines auditability and could hide malicious installation steps. Do not decode or execute the payload or install openclaw-core based on this document alone. Require an official, verifiable distribution (repository, package registry, signed binaries) and review

Confidence: 98%Severity: 75%
Audit Metadata
Analyzed At
Feb 20, 2026, 11:52 AM
Package URL
pkg:socket/skills-sh/bolhasec%2Fpost-Empirically-Testing-Skill-Scanners%2Fmoltbook-base85-obfuscation-instructions%2F@67960c57006dacfff031ea3a8667a1931619c613