launch-positioning-builder

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill's functionality is limited to market research and documentation. It uses provided agent tools for web searching and fetching according to their intended design.
  • [COMMAND_EXECUTION]: The instructions command the agent to use web_search and fetch_webpage tools to analyze competitors. It also references an external dependency on the review-scraper skill in the metadata for additional market insights.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted content from the web during its research phase.
  • Ingestion points: External competitor homepages and review sites (G2/Capterra) via the fetch_webpage tool in Phase 1.
  • Boundary markers: None present; the skill lacks specific delimiters to isolate external research data from instructional prompts.
  • Capability inventory: The agent can perform web searches, fetch webpage content, and write positioning documents to the local project directory.
  • Sanitization: There are no explicit instructions for the agent to sanitize or validate the content retrieved from external URLs before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:06 PM
Security Audit — agent-trust-hub — launch-positioning-builder