content-asset-creator
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
curlto transmit data to external generation APIs andnpx playwrightto perform the conversion of HTML assets into PDF documents. These commands are documented as core features for programmatic content creation. - [EXTERNAL_DOWNLOADS]: Styling and layout resources are fetched from well-known content delivery networks, specifically Tailwind CSS via
cdn.tailwindcss.comand Google Fonts viafonts.googleapis.com. Additionally, the use ofnpx playwrightinvolves the acquisition of browser binaries from the official npm registry. - [DATA_EXFILTRATION]: As part of its intended functionality, the skill sends user-provided content and brand settings to Gamma's API (
api.gamma.app) and Vercel's v0 platform (api.v0.dev) for processing and layout generation. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests structured data which is then interpolated into secondary LLM prompts or HTML templates. 1. Ingestion points:
Content dataandBrand configdefined in SKILL.md. 2. Boundary markers: Not specified for data interpolation. 3. Capability inventory: Execution of network requests viacurland shell commands vianpx. 4. Sanitization: No explicit escaping or validation steps are mentioned in the skill definition.
Audit Metadata