launch-x
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for the user to install additional companion skills from the author's own GitHub repository (
github.com/ulpi-io/skills). This is part of the standard functional setup for the author's toolkit. - [PROMPT_INJECTION]: The skill utilizes a web browsing tool to collect information from live product websites to inform the generation of marketing copy. This ingestion of untrusted external data creates a surface for indirect prompt injection, where malicious instructions hidden on a webpage could influence the agent's output.
- Ingestion points: Browsing of live site URLs during the grounding phase in
SKILL.md(Step 1). - Boundary markers: No explicit delimiters or boundary markers are defined to isolate external web content from the agent's internal instruction set.
- Capability inventory: The skill can write markdown files to the local project directory (
Write) and invoke companion skills (Skill). - Sanitization: There is no evidence of content validation or sanitization applied to the data retrieved from external websites before processing.
Audit Metadata