The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns or security risks were detected. The skill provides architectural guidance and best practices for context window management and the use of the Model Context Protocol (MCP) for memory persistence.
[Indirect Prompt Injection] (SAFE): The skill defines a surface for processing untrustworthy data (codebase files) to populate persistent memory. However, this is consistent with the primary purpose of a developer assistant and does not introduce unusual risk beyond the agent's baseline capabilities.
Ingestion points: File reading via Read and Grep tools, and memory retrieval via mcp__memory__search_nodes.
Boundary markers: Not explicitly defined in the documentation.
Capability inventory: The skill utilizes Bash, Write, and Edit tools for task execution.
Sanitization: Not explicitly addressed in the instructional content.

context-optimization