The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and displays untrusted data in the form of stored memories which could theoretically contain instructions designed to influence the agent.
Ingestion points: Data is retrieved from the Mem0 platform via the get_memories and search_memories tools as described in SKILL.md.
Boundary markers: The skill uses plain text indentation and dashes to structure output but does not utilize explicit boundary delimiters or 'ignore' instructions to isolate the retrieved memory content.
Capability inventory: The skill's functionality is restricted to data retrieval and formatted display to the user; it contains no file-system write operations, shell command executions, or network requests to external domains.
Sanitization: The skill implements a basic form of sanitization by truncating memory content to a maximum of 100 characters in the primary tour view, which limits the payload size of any potentially injected instructions.

mem0-tour