compdf-page-editor
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests and processes untrusted PDF documents.
- Ingestion points: Step 1 of the workflow in
SKILL.mdidentifies source PDF files for processing. - Boundary markers: The instructions lack specific delimiters or warnings to ignore potentially malicious instructions embedded within the PDF content.
- Capability inventory: The skill performs network operations (API requests to ComPDF) and file management tasks (merge, split, delete) as described in
references/endpoint-index.md. - Sanitization: There is no mention of content sanitization or validation to prevent embedded text in PDFs from influencing the agent's logic.
Audit Metadata