competitor-alternatives

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly instructs the agent to use web_search and web_fetch in Step 3 ("Research and gather differentiators" — including fetching competitor pages and review sites like G2/Capterra) and Step 7 (competitive verification), which requires reading and acting on open/public, user-generated or third‑party web content to shape drafts and decisions, so it clearly ingests untrusted third‑party content that can influence tool use and next actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires the Tiger Den connector at runtime (calls such as list_marketing_references(), get_marketing_context(slugs: ["product-marketing-context","brand-voice-guide"]), and get_marketing_reference(...)) to fetch authoritative docs whose content is injected into the agent’s prompts and thus directly controls its instructions.