The Agent Skills Directory

[SAFE]: The skill's core functionality involves analyzing and patching agent code to improve performance and behavior, which is consistent with its stated purpose as an optimizer.
[DYNAMIC_EXECUTION]: The skill generates and applies code patches and dynamically registers agents using the AGENT_REGISTRY and CAST_CODER tools as part of its optimization workflow.
[INDIRECT_PROMPT_INJECTION]: The skill features a workflow where it reads existing agent configurations and prompts to synthesize optimized versions, creating a surface for processing untrusted code.
Ingestion points: CAST_ANALYSIS.search_ast and CAST_SEARCH.read_file are used to read source files of other agents for optimization.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are mandated for the synthesized output in the provided instructions.
Capability inventory: The skill possesses the ability to write to files via CAST_CODER.search_replace and register logic via AGENT_REGISTRY.dynamic_register.
Sanitization: No explicit sanitization of the input code or content is described.
[COMMAND_EXECUTION]: The skill documentation describes using a terminal-based tool for dependency management, such as executing pip install, and performing file system operations.

optimizer