perf-ops

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

• Third-party content exposure detected (high risk: 0.80). The skill's T2 profiling and load-testing workflow (SKILL.md "Profile"/Dispatch template and references/load-testing.md and ci-integration.md) instruct agents to run HTTP load tests and profile against a provided "target application or endpoint" (e.g., k6/lighthouse examples that hit arbitrary URLs and parse JSON responses), meaning the agent will fetch and interpret untrusted/public endpoint responses which can materially influence profiling decisions and follow-up actions.