htmx-interaction-patterns
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for ingesting and rendering untrusted external data through web interfaces.
- Ingestion points: The skill provides templates for processing user input via search bars (
<input type="search" name="q">) and inline editing forms (hx-put="/skills/42") inSKILL.md. - Boundary markers: The provided code snippets do not include delimiters or instructions to ignore embedded content in the data being processed.
- Capability inventory: The skill defines network-facing endpoints and DOM manipulation patterns but does not utilize dangerous subprocess or file-system capabilities.
- Sanitization: The server-side Python snippets (e.g.,
render_results(results)andrender_update(data)) do not explicitly demonstrate sanitization or escaping of the user-provided content before it is returned as an HTML fragment to the client.
Audit Metadata