ln-812-optimization-researcher

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow (Phase 1 "Competitive Analysis" and Phase 2 "Bottleneck-Specific Research") explicitly requires using WebSearch and the MCP research tools (see shared/references/research_tool_fallback.md and references/research_query_templates.md) to fetch and ingest public URLs and web content which the agent must read and use to form hypotheses and actions, exposing it to untrusted third‑party content.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs at runtime to WebFetch files from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path} (used when shared/ is missing) and marks files like shared/references/research_tool_fallback.md as "MANDATORY READ" that determine the MCP tool priority chain and runtime behavior, meaning fetched content can directly control agent instructions.