skills/upstash/skills/upstash-box-cli/Gen Agent Trust Hub

upstash-box-cli

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could contain malicious instructions. \n
  • Ingestion points: The skill can read content from websites using box browser content and clone external repositories via box git clone. \n
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore potentially malicious embedded content within the ingested data. \n
  • Capability inventory: The skill possesses extensive capabilities including remote command execution (box exec), file modification (box files write), and automated browser interaction (box browser act). \n
  • Sanitization: There is no evidence of sanitization, filtering, or validation of content retrieved from external sources before it is processed by the agent. \n- [COMMAND_EXECUTION]: The core functionality involves executing arbitrary commands within a remote environment. \n
  • Commands such as box exec and box code enable the execution of shell scripts and programming code (Node.js, Python) in the remote container. \n- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install a specific command-line tool. \n
  • The command npm i -g @upstash/box-cli downloads and installs the necessary vendor-provided utility from the official npm registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 02:49 AM
Security Audit — agent-trust-hub — upstash-box-cli