PDF Generator
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides numerous code templates that ingest external data without implementing sanitization or boundary markers.
- Ingestion points: Several functions in
templates.md(e.g.,generate_invoice,generate_resume,generate_contract) andadvanced.md(e.g.,batch_generate,fill_form) ingest data via dictionary objects (data,field_data,data_list) meant to be filled by the agent. - Boundary markers: There are no instructions or code snippets that demonstrate the use of delimiters or 'ignore' instructions to separate untrusted data from the document structure.
- Capability inventory: The skill's code patterns facilitate file-write operations to the local filesystem using libraries such as
weasyprint,pypdf,reportlab, andfpdf2. - Sanitization: The patterns use direct string interpolation (f-strings and
.format()) to build HTML documents. This lacks escaping or validation, potentially allowing malicious data to break the HTML structure, inject unauthorized content, or trigger resource inclusion (like SSRF via<img>tags) in the generated PDF documents.
Audit Metadata