sales-jitterbit

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or high-risk vulnerabilities were identified. The skill correctly identifies and references official Jitterbit documentation and regional endpoints.\n- [PROMPT_INJECTION]: The skill implements a self-improving mechanism that appends user-provided information to references/learnings.md. This represents an indirect prompt injection surface where malicious instructions could be persisted in the agent's context. However, as this is a primary intended feature for knowledge accumulation, the risk is considered low.\n
  • Ingestion points: User-provided workarounds or tips during conversation (SKILL.md).\n
  • Boundary markers: Absent; content is appended without delimiters or instructions to ignore embedded commands.\n
  • Capability inventory: File-append access to references/learnings.md (SKILL.md).\n
  • Sanitization: Absent; the agent is instructed to append discovered information directly without validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 06:35 PM