skill-card-generator

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow runs scripts/discover_assets.py on the target skill directory and then feeds extracted file contents (including arbitrary markdown/text from that directory) into the LLM context for Jinja rendering, so any outsider-authored text in the target skill’s files can be ingested (local-file prompt injection via skill-scope content).