tracking-taxonomy-updates

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

• Third-party content exposure detected (high risk: 0.90). The skill's mandatory workflow (SKILL.md step 2: "Pull authoritative updates and release notes") and reference/sources.md explicitly list external public URLs (e.g., NCBI, GTDB, ICTV, UniEuk) that the agent is expected to fetch and interpret, meaning untrusted third-party web content can directly influence tooling and decision-making.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill includes explicit runtime commands that pull and execute remote software (e.g., docker pull / docker run of the container image bryce911/bbtools:39.65 — see https://hub.docker.com/r/bryce911/bbtools/ — and instructions to run pixi install which fetches packages from external channels documented at https://pixi.prefix.dev/), so external content will be fetched and executed as part of the skill workflow.