The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various CLI commands to manage the memory lifecycle, including memories init for project setup, memories generate to create or update AI tool configuration files such as CLAUDE.md and .cursorrules, and memories hook install to configure git post-commit hooks for automated updates.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @memories.sh/cli npm package and initiates the download of machine-learning models from remote repositories when the memories embed command is first executed.
[DATA_EXFILTRATION]: The skill is designed to synchronize local memory databases and configuration files to the memories.sh cloud service via memories login and memories sync. This involves uploading project-level and global data from directories like ~/.claude or .cursor/rules/ to a third-party server as part of its core functionality.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting rules and notes from existing files via memories ingest. 1. Ingestion points: The memories ingest command reads from local files such as CLAUDE.md and .cursorrules (File: SKILL.md, references/commands.md). 2. Boundary markers: The documentation does not specify the use of delimiters or warnings to prevent the AI from obeying instructions embedded within the ingested data. 3. Capability inventory: The skill has the ability to write to the filesystem through memories generate and transmit data over the network via memories sync. 4. Sanitization: No sanitization or validation processes are described for the content imported from external files.

memories-cli