The Agent Skills Directory

[DATA_EXPOSURE]: Accesses local configuration and state files at ~/.mem0/settings.json and ~/.mem0/mem0-dream-state.json to extract operational thresholds and scope settings.
[DATA_EXPOSURE]: Reads the MEM0_API_KEY environment variable and discloses the first six characters of the key in the status summary.
[COMMAND_EXECUTION]: Executes shell commands including grep, echo, and date to verify environment variables and parse local JSON files.
[INDIRECT_PROMPT_INJECTION]: Processes memory content retrieved from the mem0 database for quality analysis (detecting duplicates, contradictions, and tagging) without sanitization or boundary markers.
Ingestion points: Data is ingested from the memory database via get_memories and search_memories tools in SKILL.md.
Boundary markers: The skill does not define delimiters or instructions to ignore potential embedded instructions within the memory content.
Capability inventory: The skill has capabilities for shell command execution and memory management tools (add_memory, delete_memory).
Sanitization: No validation or escaping is performed on the content of the retrieved memories before the agent processes them.

mem0-status