account-tier-scoring
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted external data to drive its scoring logic and automated actions without explicit sanitization or boundary markers.
- Ingestion points: The skill ingests signal event payloads including chat interactions, outreach replies, and meeting notes (documented in
SKILL.mdStep 3 and the Signal Stack sections of both reference files). - Boundary markers: Absent. There are no instructions to delimit external content or to ignore instructions embedded within the ingested signal data.
- Capability inventory: The agent has the ability to update CRM lead scores (
SKILL.mdStep 4), enrich CRM contact records with researched data (Step 6), post to alert channels (Step 7), and draft personalized outreach (Step 7). - Sanitization: Absent. The instructions do not specify any validation, filtering, or escaping of the ingested text before it is used to influence the account tier or the content of generated outreach.
Audit Metadata