bun

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs a research sub-agent to fetch and read the public Bun documentation index (https://bun.com/docs/llms.txt) and many /docs/... pages and then use those findings to drive code/configuration and follow-up actions, so it ingests open/public third‑party content that can influence tool use and decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's research sub-agent is explicitly instructed to fetch and ingest the external documentation index at https://bun.com/docs/llms.txt at runtime, and that fetched content is used to drive the agent's outputs (controlling prompts/answers) and is a required dependency.