competitor-teardown

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Recommends installing the belt-sh/cli package via npx and references installation instructions from a GitHub repository belonging to the inference-sh organization.
  • [COMMAND_EXECUTION]: The skill uses a set of shell commands through the belt CLI to interact with research APIs and management functions.
  • [DYNAMIC_EXECUTION]: Includes a Python script template for generating a competitive positioning map, which is executed at runtime via a tool to produce a .png file.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the web, including competitor sites and user reviews, creating a surface for injection. Ingestion points: tavily/search-assistant, tavily/extract, and exa/search in SKILL.md. Boundary markers: None identified. Capability inventory: Shell access via Bash(belt *) and file system writes via Python execution. Sanitization: None observed for the fetched research data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 11:00 PM
Security Audit — agent-trust-hub — competitor-teardown