omni-embed
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions and code for handling events from an embedded iframe, which introduces a potential attack surface for indirect prompt injection if the agent processes untrusted data from the iframe without proper validation.\n
- Ingestion points: Client-side event handling using
window.addEventListener('message', ...)as described inSKILL.md.\n - Boundary markers: The skill includes explicit instructions and code snippets to validate the source of messages (
event.data?.source === 'omni') before processing.\n - Capability inventory: The skill allows the agent to generate signed URLs and execute shell commands via the Omni CLI (
omni scim,omni documents).\n - Sanitization: While message source validation is provided, the skill focuses on integration logic and assumes standard application-level sanitization for specific event payloads.\n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@omni-co/embedpackage from the NPM registry. This is a legitimate vendor resource associated with the Omni platform.\n- [COMMAND_EXECUTION]: The skill utilizes theomniCLI for administrative tasks, user lookups, and document discovery, which is consistent with its stated purpose of managing dashboard embedding.
Audit Metadata