enterprise-clone-builder

Warn

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell scripts to perform web scraping via an external tool.\n
  • Evidence: references/web-clipper-usage.md provides bash commands to execute scripts located at $WEB_CLIPPER_ROOT/scripts/run_web_clipper.sh for single and batch URL processing.\n- [EXTERNAL_DOWNLOADS]: The skill initiates network requests to retrieve data from arbitrary external sources.\n
  • Evidence: SKILL.md (Step 1) and references/web-clipper-usage.md describe the automated crawling of official enterprise websites and third-party platforms to gather news, products, and brand information.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing untrusted data to generate system prompts.\n
  • Ingestion points: Local files (PDF, Word, Excel, PPT) and scraped web content as specified in SKILL.md (Step 0.2 and Step 1).\n
  • Boundary markers: Absent. There are no delimiters or instructions to ignore malicious commands that might be embedded in the processed materials.\n
  • Capability inventory: The skill writes a core identity file (AGENTS.md) containing system prompts and executes shell scripts for web scraping.\n
  • Sanitization: Absent. Data is extracted from raw source files and used directly to build structured profiles and agent instructions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 23, 2026, 05:13 AM
Security Audit — agent-trust-hub — enterprise-clone-builder