The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill requires downloading, installing dependencies, and compiling source code from the community repository webzler/agentMemory via npm install and npm run compile.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by automatically importing data from existing project markdown files.
Ingestion points: The workflow attempts to import memory banks from files located in .kilocode/, .clinerules/, and .roo/ directories.
Boundary markers: The instructions do not specify the use of delimiters or explicit warnings to ignore embedded instructions within the ingested data.
Capability inventory: The skill provides tools like memory_search, memory_write, and memory_read which directly influence the agent's context and decision-making history.
Sanitization: There is no documentation regarding the sanitization or filtering of content imported from external project files.

agent-memory