spec-research

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow explicitly includes “Investigate external sources: documentation, APIs, library comparisons, platform constraints” and “Read existing project context … other research/*.md” which can include outsider-authored content (e.g., public web pages fetched at runtime or issue/wiki/discussion text delivered via APIs) that the agent may ingest as readable prose into the LLM context.