vllm-pytorch-ci-triage

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text is ingested when the skill fetches Buildkite job logs from https://api.buildkite.com/.../jobs/<N>/log.txt (public CI output authored by other parties/runner processes) and then strips/feeds that log text into the agent’s LLM context for root-cause extraction.