baoyu-post-to-x

[Skill Scanner] Backtick command substitution detected All findings: [HIGH] command_injection: Backtick command substitution detected (CI003) [AITech 9.1.4] [HIGH] command_injection: Backtick command substitution detected (CI003) [AITech 9.1.4] [HIGH] command_injection: Backtick command substitution detected (CI003) [AITech 9.1.4] The skill description and declared capabilities are internally consistent with an automation tool that uses a real browser to post to X. There are no direct signs of obfuscation, hardcoded credentials, or third-party exfiltration in the provided SKILL.md. The primary security concerns are operational: running these scripts grants the automation access to the logged-in Chrome profile, filesystem-specified content (images, markdown), and clipboard — which could be misused if the scripts are tampered with or run on a compromised machine. Before use, verify the actual scripts in scripts/*.ts, ensure you trust the source, and run in a controlled profile or test account to limit impact. LLM verification: This SKILL.md describes a legitimate-looking browser-automation skill to post to X that has an appropriate purpose-capability alignment on its surface. However, the operational model requires high privileges (real browser control, optional reuse of Chrome profile, clipboard and keystroke injection) and includes an extension-loading mechanism (EXTEND.md) and run-time invocation patterns (npx -y bun) that increase supply-chain and credential-exposure risk. There is no direct evidence of obfuscatio