The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and index data from the local project workspace, including markdown files and specific configuration directories such as .kilocode/, .clinerules/, and .roo/. * Ingestion points: The skill reads project documentation and rule files directly into its memory bank. * Boundary markers: There are no explicit instructions or delimiters mentioned to prevent the agent from executing instructions potentially embedded within these ingested files. * Capability inventory: The skill possesses tools for file reading/writing (memory_read, memory_write) and requires shell command execution for its operational server. * Sanitization: The skill description does not specify any sanitization or validation of the ingested markdown content.
[COMMAND_EXECUTION]: The setup and operation of the skill require the execution of shell commands. Evidence: Instructions include running 'npm install', 'npm run compile', and 'npm run start-server' with workspace path arguments.
[EXTERNAL_DOWNLOADS]: The 'npm install' command triggers the download of third-party dependencies from the public NPM registry. Evidence: The installation step follows standard Node.js practices for a community-provided repository.

agent-memory