skill-card-generator

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

• Third-party content exposure detected (high risk: 0.85). The required runtime path runs scripts/discover_assets.py on the target skill directory, which reads and emits skill/repo file contents (including arbitrary markdown/text authored by outsiders) into the agent’s context JSON and then into the Jinja-rendered LLM prompt context; this is an outsider free-text ingestion vector via local repo files (e.g., issue/wiki/PR text or any third-party-authored docs within the target skill directory).