The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for standard Node.js project initialization, including npm install, npm run compile, and npm run start-server. These are used to build and run the local memory server.
[DATA_EXPOSURE]: The skill accesses the local project workspace to manage knowledge, specifically reading from and writing to directories like .agentMemory, .kilocode/, .clinerules/, and .roo/. This behavior is consistent with the skill's stated purpose of providing persistent, searchable memory for the project.
[PROMPT_INJECTION]: As a knowledge management tool, the skill has an attack surface for indirect prompt injection, where malicious instructions embedded in project documentation could be stored and later retrieved to influence agent behavior.
Ingestion points: The skill imports markdown content from project directories like .kilocode/, .clinerules/, and .roo/ (SKILL.md).
Boundary markers: No specific delimiters or "ignore instructions" warnings are mentioned for the retrieved memory content.
Capability inventory: The skill provides tools for searching, reading, and writing local memory data (memory_search, memory_read, memory_write).
Sanitization: No sanitization or validation of the stored memory content is described in the documentation.

agent-memory