document-ocr

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly instructs the agent to ask the user for the backend host/model or API key and includes examples of exporting/passing API keys (e.g., export OPENAI_API_KEY, --api-key), so the agent may need to receive and could end up echoing secret values verbatim, creating an exfiltration risk.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)