agent-app-store

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's example explicitly shows secrets (api_key and database_url) passed inline as variables to the deployment API, which requires the agent to handle and emit secret values verbatim in calls/outputs.

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow includes marketplace browsing/publishing/deployment via MCP calls (e.g., mcp__flow-nexus__app_search and mcp__flow-nexus__app_store_publish_app) that can ingest app metadata/content authored by third-party developers into the agent’s LLM context (outsider-authored free text via marketplace records).