The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes data fetched from an external REST API (Omni Analytics), which introduces a surface for indirect prompt injection if the API response contains adversarial instructions.
Ingestion points: Responses from the Omni query API endpoints as described in SKILL.md.
Boundary markers: Absent; there are no specific instructions for the agent to isolate or treat data from the API as untrusted content.
Capability inventory: The skill utilizes network access via curl and data manipulation using Python snippets in SKILL.md.
Sanitization: Absent; the instructions do not mention validating or filtering the data retrieved from the API.
[EXTERNAL_DOWNLOADS]: The skill references standard data analysis libraries to handle query results.
pyarrow and pandas are mentioned in SKILL.md for decoding and processing Apache Arrow data.
[COMMAND_EXECUTION]: The skill provides example CLI commands for interacting with external services.
Uses curl to fetch OpenAPI specifications and execute query requests against Omni Analytics endpoints.

omni-query