cognee-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill instructs use of cognee-cli config set <key> <value> (writing secrets into .env) and exposes --show-secrets, which encourages passing and displaying API keys as literal CLI arguments/outputs, an insecure pattern that would require the LLM to handle/output secret values verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow (cognee-cli remember/add) ingests user-provided free text/paths/URLs, then builds the graph via add + cognify, so outsider-authored content can be processed directly without selecting a specific trusted item first.