The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, creating a surface for indirect prompt injection.
Ingestion points: The tools ctx_execute for API responses, ctx_execute_file for local data files (such as CSV, JSON, and logs), and ctx_fetch_and_index for external web documentation.
Boundary markers: No explicit boundary markers or instructions to ignore embedded instructions are provided for the data being processed.
Capability inventory: The skill utilizes ctx_execute and ctx_execute_file to run arbitrary JavaScript, Python, and Shell code, and includes standard Bash tools for file and system operations.
Sanitization: No explicit sanitization or validation steps are defined for the ingested content before it is summarized and returned to the agent context.
[COMMAND_EXECUTION]: The skill provides extensive code templates for JavaScript, Python, and Shell to be executed dynamically via the ctx_execute and ctx_execute_file tools for the purpose of data analysis and summarization.
[EXTERNAL_DOWNLOADS]: Guidelines include instructions for fetching data and documentation from external URLs, including GitHub repositories and arbitrary API endpoints.

context-mode