The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx @arizeai/phoenix-cli to download and execute the CLI tool directly from the npm registry. This is a standard distribution method for Node.js-based vendor tools.
[COMMAND_EXECUTION]: The skill relies on shell command execution for all its functionality, including resource management (trace, span, dataset) and documentation retrieval (px docs fetch).
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes trace data, span attributes, and dataset examples that may contain untrusted content from previous LLM interactions.
Ingestion points: External data is ingested through commands like px trace get, px span list, and px dataset get in SKILL.md.
Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands within the fetched trace/span data.
Capability inventory: The agent has the capability to execute shell commands via the px CLI and process the output.
Sanitization: No sanitization or validation of the retrieved trace/dataset content is implemented before it is presented to the agent context.

phoenix-cli