frontend-agents-a11y-perf-consistency-auditor

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow reads and audits arbitrary in-scope *.html/*.css/*.js/*.ts/*.md files from the provided scope (e.g., a pull request diff or newly added files), so any outsider-authored free text contained in those files can be ingested into the agent’s LLM context via the “Read source materially” step.