posthog-automation

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow uses PostHog MCP tools (e.g., POSTHOG_LIST_AND_FILTER_PROJECT_EVENTS, POSTHOG_LIST_AND_MANAGE_PROJECT_FEATURE_FLAGS) which return event/flag data as readable text/JSON from the PostHog tenant; this content is not authored by the operating user and can include outsider-provided freeform properties (e.g., event properties), which the agent ingests into LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill requires adding and querying the MCP endpoint https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS) to fetch current tool schemas that directly determine agent behavior and available instructions, so the external URL can control prompts at runtime.