sales-execution

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions and scripts use the official hubspot CLI along with common shell utilities such as jq, date, paste, and cat. It manages data flow using temporary files in the /tmp directory (e.g., /tmp/deal_tasks.jsonl).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from an external CRM system, creating a potential surface for indirect prompt injection if the data contains malicious instructions.
  • Ingestion points: Data is retrieved from the CRM via the hubspot objects search, hubspot associations list, and hubspot objects get commands in SKILL.md.
  • Boundary markers: The skill does not define specific boundary markers or instructions for the agent to ignore embedded commands within the retrieved CRM data.
  • Capability inventory: The skill's capabilities include creating CRM objects (hubspot objects create), managing associations (hubspot associations create), and writing to local temporary files.
  • Sanitization: While data is structured using jq, there is no explicit validation or sanitization of user-controlled text fields (like hs_call_body or hs_task_subject) before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 09:51 PM