The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (the Onyx knowledge base) via the search and ask commands. This represents a potential indirect prompt injection surface if the connected data sources (Slack, Google Drive, Confluence, etc.) contain malicious instructions intended to influence the agent's behavior when the content is retrieved.
Ingestion points: Data enters the agent's context through the output of onyx-cli search and onyx-cli ask commands.
Boundary markers: The search command provides structured JSON output, which helps delimit data from instructions. The ask command provides plain text.
Capability inventory: The agent can execute onyx-cli subcommands and process the returned results.
Sanitization: The instructions do not specify any sanitization or filtering of the retrieved content.
[EXTERNAL_DOWNLOADS]: The documentation includes instructions to install the onyx-cli package from PyPI (pip install onyx-cli). This is the primary tool required for the skill's functionality and is provided by the same vendor as the skill author.

onyx-cli