parallel-search

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The docs and root URLs look like ordinary documentation pages, but the explicit https://parallel.ai/install.sh and the curl -fsSL https://parallel.ai/install.sh | bash install pattern is a high-risk distribution vector (remote shell script execution) from an unverified domain, so it could be used to deliver malware unless you verify the script and publisher.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to use the Parallel web search API to fetch and return excerpts from open/public sites (results[].excerpts, URLs, titles) — including examples like Reddit/GitHub/news — which the agent is expected to read and synthesize into answers, exposing it to untrusted third‑party content that could carry indirect prompt injections.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's Prerequisites explicitly instruct running "curl -fsSL https://parallel.ai/install.sh | bash", which fetches and executes remote code from https://parallel.ai/install.sh and is required to install the CLI used by the skill.