sharepoint-safe-html
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from user requests and external SharePoint-hosted files (CSV and XLSX) to generate HTML artifacts. This creates an attack surface for indirect prompt injection where malicious content in those files could attempt to influence the agent's behavior or the generated output. * Ingestion points: User prompts and attached data files; SharePoint-hosted .csv and .xlsx files referenced for live-linking (described in SKILL.md Step 2A). * Boundary markers: The skill does not define explicit delimiters for untrusted data in its prompts, relying instead on structural instructions and a validation checklist. * Capability inventory: The skill generates functional HTML and inline JavaScript, providing a mechanism for content to be rendered in a user's browser. * Sanitization: Step 5 of the workflow provides a mandatory HTML escaping table for all dynamic content.
- [COMMAND_EXECUTION]: The skill is capable of generating inline JavaScript for interactivity. Although it explicitly prohibits the use of eval(), dynamic script injection, and external dependencies, the automated generation of executable scripts based on user-provided data remains a security consideration.
Audit Metadata