community-moderation-brief
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially malicious external content such as comments, DMs, and Discord posts, creating a surface for indirect prompt injection. * Ingestion points: Item, channel, and thread summaries from creator communities as defined in SKILL.md and references/community-moderation-brief-checklist.md. * Boundary markers: The instructions include specific mandates to keep systems 'read-only' unless approved, and requirements to route safety or legal items to human reviewers. * Capability inventory: The skill contains no executable scripts, tool definitions, or network access commands. * Sanitization: The workflow requires summarizing issues without overstating intent and separates internal notes from public responses to prevent accidental execution of embedded instructions.
Audit Metadata