hmos-memleak-analysis

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt requires reading source files and including verbatim code snippets in the generated report (and running analysis scripts on the repo), so if secrets (API keys, tokens, passwords) exist in the code they would be output directly, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow runs scripts/filter_on.py/filter_risk_func.py over the user-supplied target source code dirpath, and those scripts read and embed the scanned source files’ code lines (outsider-authored project files) into the LLM/agent context via the generated “skill”/report fields like code_snippet/code.