The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of an external Python package blockrun-llm from a public registry to function.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for installing and updating the SDK using pip install blockrun-llm and pip install --upgrade blockrun-llm.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, specifically real-time X (Twitter) and web search results via the Grok/Live Search integration.
Ingestion points: Data enters the context through client.chat() calls when search=True is enabled in SKILL.md.
Boundary markers: None identified in the provided instructions; external content is directly integrated into the agent's reasoning process.
Capability inventory: The skill allows file reading, Python execution, and network operations through the SDK to facilitate image generation and model routing.
Sanitization: No explicit sanitization or filtering of the external search results is described in the skill instructions.

blockrun