cold-email

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of Markdown files providing copywriting frameworks and benchmarks.
  • [DATA_EXPOSURE]: The skill instructs the agent to read context from specific workspace files (.agents/product-marketing-context.md), which is part of the intended workflow and does not target system secrets.
  • [INDIRECT_PROMPT_INJECTION]: The agent is instructed to ingest external data such as LinkedIn posts or company news provided by the user. While this creates a potential surface for indirect prompt injection, the skill lacks any capabilities (network, shell, file writes) to facilitate an attack.
  • Ingestion points: SKILL.md (User-provided research signals).
  • Boundary markers: Absent.
  • Capability inventory: None. No scripts or executable actions are defined in this skill.
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 07:41 AM