literature-review

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most URLs are reputable academic and publisher resources, but the presence of a direct shell installer URL (https://parallel.ai/install.sh) and the curl | bash installation pattern in the skill prompt are high‑risk indicators because remote scripts/executables can deliver malware if not from a verified, trusted source.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses parallel-cli search / parallel-cli extract to fetch and ingest full text/abstracts from public web sources (e.g., arbitrary URLs from scholarly domains), which are outsider-authored free text that can be included into the agent’s LLM context for screening/extraction.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires installing the primary runtime tool via a remote install script that is fetched and executed (curl -fsSL https://parallel.ai/install.sh | bash), which downloads and runs remote code as a required dependency for the skill.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)