momentic-result-classification

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow (SKILL.md) requires calling momentic_get_run / momentic_list_runs to load run result folders from the MCP working directory (e.g., .momentic-mcp) and explicitly directs the agent to inspect assets such as .html and screenshots from past runs, which are untrusted/third-party DOM snapshots and can materially influence classification and subsequent actions.