PDF Generator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill processes untrusted Markdown and HTML data to generate PDFs, creating a surface for injection attacks.
- Ingestion points:
markdownToPdffunction reads external files from themdPathparameter. - Boundary markers: Absent. There are no delimiters or warnings to the agent about ignoring instructions embedded in the source documents.
- Capability inventory: The skill uses
puppeteer(process spawning),fs.readFileSync(file read), andfs.writeFileSync(file write). - Sanitization: Absent. The code uses
marked.parse(markdown)and immediately renders the output viapage.setContent(styledHtml). Maliciously crafted Markdown containing<script>tags or<iframe>elements could lead to local file disclosure or SSRF within the headless browser context. - Command Execution (LOW): The skill invokes
puppeteer.launch(), which spawns a Chromium subprocess. While this is the intended primary purpose of the skill, it represents a standard risk factor for executing code in a browser environment that processes untrusted inputs.
Audit Metadata