obsidian-second-brain

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). Most links are GitHub/Gist resources (including a reputable Karpathy gist) and benign content (YouTube/example), but the set includes direct raw.githubusercontent.com shell-install scripts and instructions to curl|bash or npx-install code from third-party GitHub accounts (akillness, eugeniughelbur), which is a high-risk pattern because running unvetted remote scripts or installing executables from unknown repos can deliver malware.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow includes /obsidian-ingest <url|file|image|audio> which fetches/reads outsider-provided web content (public pages) or outsider-authored media/transcripts, then rewrites/saves that readable text into the vault/LLM context—an indirect prompt-injection exposure path.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's install instructions explicitly fetch-and-run remote code (e.g., "curl -fsSL https://raw.githubusercontent.com/akillness/obsidian-second-brain/main/scripts/quick-install.sh | bash" and the git clone of https://github.com/akillness/obsidian-second-brain followed by "bash .../scripts/setup.sh"), which downloads and executes external scripts during setup and therefore can directly execute code that controls the agent.