post-scorer
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses imperative instructions ("CRITICAL: Auto-start on load", "Do not summarise") to override the agent's standard conversational behavior and force immediate execution of the scoring logic.
- [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection as it processes untrusted external data (LinkedIn posts fetched via Apify) and user-supplied drafts to perform its analysis. * Ingestion points: User-provided post text and historical data retrieved through the apimaestro/linkedin-profile-posts tool. * Boundary markers: No delimiters or warnings are used to prevent the agent from obeying instructions that might be embedded in the analyzed posts. * Capability inventory: The skill performs file read and write operations for caching engagement data and makes external network calls via tools. * Sanitization: No sanitization or filtering logic is present for the ingested content.
Audit Metadata