tapestry

[Skill Scanner] Installation of third-party script detected All findings: [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] This SKILL.md script is consistent with its stated purpose: detecting URL content type, extracting content, and producing an action plan. I found no evidence of credential harvesting, third-party proxying of user content, obfuscation, or embedded backdoors. The primary security concerns are operational: it downloads and processes arbitrary URLs and may auto-install tools (brew install yt-dlp). Running it against untrusted or unknown URLs (or on an automated system) could expose the host to vulnerabilities in external tools or crafted content. Use with caution: validate URLs, run in a sandbox, and avoid auto-install behavior in sensitive environments. LLM verification: This AI agent skill's behavior is consistent with its stated purpose (detect URL type and extract content). It does not contain obvious obfuscated or clearly malicious code. However, it executes external tools, downloads arbitrary URLs, runs inline Python on fetched content, and installs packages at runtime via system package managers without integrity checks — these actions raise supply-chain and execution risks. Recommend treating this skill as potentially unsafe to run in sensitive or privile