bootstrap-docs

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md (Initialization step 3 and the Appendix "TEMPLATE_BASE_URL") explicitly directs the agent to fetch templates from a public GitHub raw URL (https://raw.githubusercontent.com/nesnilnehc/project-documentation-template/main/), which the agent ingests and uses as the authoritative target reference to drive structural decisions and apply changes.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches template files at runtime from https://raw.githubusercontent.com/nesnilnehc/project-documentation-template/main/ (and references https://github.com/nesnilnehc/project-documentation-template), and those fetched templates/ADR files are injected/used to guide the agent's document-generation instructions, so the external content can directly control agent prompts.