huawei-cloud-msmodelslim-model-analysis

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can enter via runtime download of non-weight files from ModelScope using a user-supplied <org>/<model> identifier; the skill then reads config.json and modeling_*.py from that downloaded directory, which are outsider-authored text files ingested into the LLM context during analysis.

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

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