literature-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill’s Semantic Scholar command injects an API key via a --api-key command-line argument (using shell substitution from a local file), which requires embedding the secret into a generated command and exposes it via process args/command history, so it’s an insecure handling pattern.

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow and included scripts explicitly fetch and parse public, untrusted content—e.g., SKILL.md directs running Semantic Scholar/OpenAlex/CrossRef searches and scripts/search_openalex.py and scripts/search_crossref.py call public APIs, and scripts/download_arxiv_source.py downloads and extracts arXiv .tex source—so the agent will read/interpret third-party (user-generated/preprint) content that could influence its decisions or actions.