The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for processing user-supplied text to generate embeddings through the OmniRoute service, creating a surface for potential instruction injection in the ingested data.
Ingestion points: User text is accepted via the input field in the embedding request examples in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore potential commands within the input text are included.
Capability inventory: The skill utilizes network access to transmit data to the external $OMNIROUTE_URL endpoint.
Sanitization: No sanitization or validation of the input text is mentioned.
[SAFE]: Configuration and authentication are securely managed using environment variables (OMNIROUTE_URL, OMNIROUTE_KEY), preventing hardcoded credential exposure.
[SAFE]: The skill references external documentation on GitHub for setup guidance, which is standard practice and does not involve automated remote code execution.

omniroute-embeddings