signals-scout-customer-analytics

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes data from the system.accounts and events tables, creating a surface for indirect prompt injection where malicious content in database fields could influence the agent's behavior. \n- Ingestion points: The skill retrieves account names and properties (e.g., csm, account_executive) via SQL queries in SKILL.md. \n- Boundary markers: The skill does not define specific delimiters or isolation techniques to separate untrusted data retrieved from the database in SKILL.md. \n- Capability inventory: The skill has access to tools for executing SQL, writing to persistent memory (signals-scout-scratchpad-remember), and emitting signals to users (signals-scout-emit-signal) in SKILL.md. \n- Sanitization: No sanitization or validation of database content is performed before the data is used in analysis or included in signals in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 09:49 AM
Security Audit — agent-trust-hub — signals-scout-customer-analytics