social-card-gen
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The skill processes untrusted data from local files via the --file argument, remote URLs via the --url argument, and raw text inputs.
- Boundary markers: Neither the manual instructions in SKILL.md nor the generate.js script utilize explicit boundary markers or instructions to disregard embedded commands in the source message.
- Capability inventory: The skill includes scripts with the ability to read from the local file system (fs.readFileSync), write to the local file system (fs.writeFileSync, fs.mkdirSync), and make network requests (fetch).
- Sanitization: The generate.js script implements a stripMarkdownAndHtml function that removes code blocks, HTML tags, and markdown formatting, providing basic structural sanitization.
Audit Metadata