sciverse-academic-retrieval

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly requires providing an SCP Hub API key and shows it being passed directly into headers (SCP-HUB-API-KEY) and as an inline API_KEY variable in example code, which encourages embedding secret values verbatim in generated code or commands.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow calls MCP tools like semantic_search and read_content, whose returned chunk/text are extracted as readable text and would be fed into the agent’s LLM context; that content is outsider-authored scientific literature (public papers/preprints) and can include arbitrary free text.