media-procurement-benchmark

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

• Third-party content exposure detected (high risk: 0.75). The required workflow includes a mandatory runtime web search for benchmarks (public web content fetched at runtime) and then passes the searched results into the agent/LLM context via benchmark_overrides for citation and scoring, creating an indirect prompt-injection exposure path.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)