competitor-teardown

Warn

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and directs the user to fetch installation instructions from an external GitHub repository (inference-sh/skills).
  • [REMOTE_CODE_EXECUTION]: Instructions direct the user to install a third-party CLI tool (belt-sh/cli) via npx, which executes code from an external registry.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection.
  • Ingestion points: External data is ingested via tavily/search-assistant, tavily/extract, exa/search, and infsh/agent-browser (SKILL.md).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing flow.
  • Capability inventory: The skill is allowed to use Bash(belt *), providing access to a suite of search, browsing, and execution tools (SKILL.md).
  • Sanitization: No evidence of sanitization, filtering, or validation of the retrieved web content is provided.
  • [DYNAMIC_EXECUTION]: The skill uses a Python execution tool (infsh/python-executor) to run code for generating positioning maps, involving runtime execution of script templates (SKILL.md).
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 7, 2026, 10:15 AM
Security Audit — agent-trust-hub — competitor-teardown