competitor-teardown

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the belt CLI to run various market research apps, including search assistants and browser agents for capturing screenshots.
  • [EXTERNAL_DOWNLOADS]: Includes instructions for installing the belt CLI from the inference-sh organization and suggests adding related skills for web searching and prompt engineering.
  • [REMOTE_CODE_EXECUTION]: Employs the python-executor app to run a matplotlib script that generates a positioning map image. The script is static and serves a documented visualization purpose.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the open web via search and extraction tools.
  • Ingestion points: Results from Tavily search, Exa search, and direct website extraction.
  • Boundary markers: Absent; there are no explicit delimiters or warnings to ignore instructions within the fetched data.
  • Capability inventory: The skill has access to shell commands (Bash) and Python execution (python-executor).
  • Sanitization: No sanitization or filtering of external content is performed before the data is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 07:38 PM
Security Audit — agent-trust-hub — competitor-teardown