hyperframes-website-to-video

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most links are to well-known APIs/CDNs and documentation, but the GitHub repo https://github.com/TheAngryPit/TheAngrySkills.git is a third‑party skill collection used by the install command in this pack and represents a supply‑chain risk if the author/account is untrusted.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The repo intentionally redirects installs/updates to a mirrored third-party pack and includes explicit guidance to accept/paste API keys in chat and run a "silent" install command — patterns that enable supply-chain control and credential capture/exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text is ingested via the runtime website capture path (npx hyperframes capture <URL> in Step 0), which downloads the user-supplied public URL and extracts readable text like capture/extracted/visible-text.txt / asset-descriptions.md that later steps read into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime commands and runtime script imports that fetch and execute remote code — notably the npx install pointing at https://github.com/TheAngryPit/TheAngrySkills.git and CDN-loaded libraries like https://cdn.jsdelivr.net/npm/gsap@3.14.2/dist/gsap.min.js and https://cdn.jsdelivr.net/npm/three@0.181.2/+esm which are fetched/loaded at runtime and thus execute remote code required by the skill.