macaron-card-generator
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The
scripts/generate_card.pyscript incorporates user-influenced text directly into HTML templates without escaping. This creates a surface for indirect prompt injection, as malicious input could inject HTML or JavaScript into the rendered card. - Ingestion points: Text content provided by the user or external sources which is passed to the
--contentargument of the generator script. - Boundary markers: Absent; there are no instructions or code mechanisms to validate or escape the input strings.
- Capability inventory: Execution of
scripts/generate_card.py, file system writing, and rendering via browser tools (Playwright). - Sanitization: Absent; the Python script uses f-string interpolation for all content fields directly into the HTML and CSS blocks.
- [COMMAND_EXECUTION]: The workflow relies on executing a local Python script and browser-based rendering tools to process potentially untrusted input. This multi-step process represents a potential execution chain for injected content if inputs are not properly sanitized.
Audit Metadata