spec-writing

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md’s required workflow instructs the agent to “Read the writing rules and abstract-verb vocabulary selected by the Language Rules” (Step 2), which at runtime ingests the contents of references/abstract-verbs*.md and references/writing-rules*.md; these reference files are sourced from an outsider repository/gist (k16shikano/japanese-tech-writing) and thus constitute outsider-authored free text.