The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructions direct the user to clone an external repository (https://github.com/webzler/agentMemory.git) and execute its content using npm run start-server and npm run start-dashboard. The source repository is not managed by a trusted organization or the skill author.
[EXTERNAL_DOWNLOADS]: The setup process requires fetching code from a third-party GitHub account during initialization.
[COMMAND_EXECUTION]: The skill executes multiple shell commands (npm ci, npm run compile, npm run start-server) that trigger lifecycle scripts defined in the external repository's package.json, which could lead to arbitrary code execution on the host machine.
[INDIRECT_PROMPT_INJECTION]: As a memory management system, this skill creates a persistent data store that the agent reads from and writes to. This represents an attack surface where untrusted data saved to memory could be used to influence the agent's future logic or instructions.
Ingestion points: Data is ingested through the memory_write tool.
Boundary markers: None identified in the skill instructions to separate memory content from system instructions.
Capability inventory: The skill has the ability to execute code via the MCP server setup.
Sanitization: No explicit sanitization or validation of the memory content is defined.

agent-memory-mcp