app-platform-migration

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow ingests outsider-authored free text from the user-provided repository (e.g., Procfile, app.json, heroku.yml, docker-compose.yml, Dockerfile, and other scanned source files) by cloning/reading them and then passing their extracted contents/derived strings into the agent’s LLM context for analysis and spec generation.