analyzing-malware-family-relationships-with-malpedia

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The code builds an Authorization header from an api_key and instantiates MalpediaClient with a literal "YOUR_MALPEDIA_API_KEY", which instructs embedding an API key verbatim in output/code — an insecure secret-handling pattern.

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

• Third-party content exposure detected (high risk: 0.75). The required workflow queries the Malpedia API at runtime and ingests its JSON fields (e.g., description, alt_names, attribution, and potentially YARA rule text) into the agent’s execution context, which is outsider-authored free text from a public third-party dataset.