The Agent Skills Directory

[DATA_EXPOSURE]: The skill reads the application's configuration file located at ~/.claude-mem/settings.json to identify the active worker port. This is a legitimate operation for managing the specific application's state.
[COMMAND_EXECUTION]: The skill uses curl to interact with the local worker API at 127.0.0.1 for status checks and service restarts. It also provides instructions for the user to run npx claude-mem install in their own terminal.
[DYNAMIC_EXECUTION]: A node -e command is used to programmatically parse the local settings file and resolve the worker port dynamically.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from a local API endpoint (/api/sync/status).
Ingestion points: Worker status route via curl in SKILL.md.
Boundary markers: None identified in the script logic.
Capability inventory: Uses Bash tool for local API interaction and Read for configuration access.
Sanitization: No explicit sanitization of API output before reporting.

cloud-sync