instantly-sequences

Pass

Audited by Gen Agent Trust Hub on Oct 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a local utility script, scripts/check-output.py, which is used to validate that generated deliverables contain required sections and terminology. The script performs basic text analysis and regex matching.
  • [COMMAND_EXECUTION]: Documentation in references/clay-enrollment-handoff.md references the use of leadmagic-cli (via lm commands) for automating lead data transfers. These commands represent vendor-owned tools associated with the skill's author and are used for their intended integration purposes.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting external data from CSV files, enrichment tables (Clay), and webhooks. This establishes an attack surface for indirect prompt injection, particularly within personalization fields like why_now or hook_line in lead lists. However, the skill incorporates validation steps (e.g., email verification gates) and the data processing is fundamental to the skill's utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 10, 2026, 12:53 PM