platform-assistant

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 fetches plain-text portal files from a public Anypoint Platform developer portal (e.g., GET {portal-url}/llms.txt, AGENTS.md, registry.json, and skills/{slug}/SKILL.md), which are outsider-authored content that can include free-form text; that fetched text is then ingested into the agent’s LLM context for parsing/execution guidance.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). This skill explicitly requires fetching runtime portal files (e.g., {portal-url}/registry.json and {portal-url}/skills/{slug}/SKILL.md) which contain agent-executable workflows and instructions that directly control the agent's prompts and execution, so these URLs are runtime external dependencies that control the agent.