document-accessibility-wizard
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
findutility to discover and inventory documents in the user's filesystem. - [DYNAMIC_EXECUTION]: The skill generates remediation scripts (Bash, PowerShell) and CI/CD configuration files (GitHub Actions, Azure DevOps) based on audit results.
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from untrusted external documents (Office files and PDFs) via sub-agents. This creates an attack surface for indirect prompt injection where malicious content within a document could attempt to influence the audit report or agent behavior.
- Ingestion points: Document files (.docx, .xlsx, .pptx, .pdf) processed in SKILL.md Phase 1 and Phase 2.
- Boundary markers: The skill uses a structured 'Document Scan Context' block to pass information to sub-agents.
- Capability inventory: The skill uses
findfor file discovery and generates shell scripts, markdown reports, and CSV/JSON exports. - Sanitization: The skill does not describe specific sanitization of external document content before processing.
- [SAFE]: The skill references authoritative accessibility standards from reputable sources such as the W3C and PDF Association.
Audit Metadata