community-management
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a workflow where the agent must ingest and respond to untrusted data from external users (follower comments, direct messages, and mentions), creating a surface for indirect prompt injection.
- Ingestion points: Process described in Step 1, Section 3 ('Daily Engagement Routine') involves checking overnight comments, replying to DMs, and monitoring mentions/tags.
- Boundary markers: Absent. There are no instructions to use delimiters or ignore embedded instructions within user-provided content.
- Capability inventory: The skill requires the agent to generate and send outbound messages (replies, direct messages, and community posts).
- Sanitization: Absent. The guidelines do not include validation or filtering steps for the ingested user content before it is processed by the agent.
Audit Metadata