sales-getprospect

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill gathers context from user input regarding their roles and business goals to provide tailored guidance, which creates an indirect prompt injection surface.
  • Ingestion points: User input gathered in Step 1 (role, goals, and platform area) and historical knowledge read from references/learnings.md.
  • Boundary markers: Absent. The skill does not employ specific delimiters or instructions to isolate user-supplied context from its own system prompts.
  • Capability inventory: The skill is primarily instructional and does not include subprocess execution, arbitrary code execution, or direct network operations; it does, however, instruct the agent to write updates to references/learnings.md.
  • Sanitization: Absent. There is no evidence of validation or filtering of the data ingested from user inputs before it is processed or stored.
  • [NO_CODE]: The skill consists entirely of Markdown instructions and documentation reference files with no accompanying scripts or binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 06:26 PM