launchfast-full-research-loop
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection vulnerability surface identified.
- Ingestion points: The skill ingests untrusted data from external e-commerce and intellectual property databases via the
mcp__launchfast__research_products,mcp__launchfast__ip_check_manage,mcp__launchfast__supplier_research, andmcp__launchfast__amazon_keyword_researchtools (referenced in SKILL.md). - Boundary markers: Absent. External data retrieved by tools is interpolated directly into the HTML report template without delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill possesses the capability to write files to the user's filesystem (specifically
~/Downloads/) and performs network operations via the associated MCP tools. - Sanitization: None detected. The skill does not include instructions to sanitize, escape, or validate the content retrieved from external sources before rendering it into the final HTML document.
- [COMMAND_EXECUTION]: The skill orchestrates multiple operational phases using specialized MCP tools (LaunchFast vendor resources). These tools execute data retrieval and analysis tasks across various external platforms (Amazon, Alibaba, trademark databases) to produce the research results.
Audit Metadata