The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (parse_export_file.py and import_competing_tools.py) located within the plugin's root directory. These scripts are used to process and parse data from various files.
[PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection (Category 8) because it ingests and processes data from external sources that may be attacker-controlled.
Ingestion points: Processes data from Markdown export files (mem0-export*.md), third-party AI tool configurations (.cursorrules, .github/copilot-instructions.md, .continue/rules.md), and Claude Code's native MEMORY.md file.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the imported files.
Capability inventory: The skill possesses the capability to execute shell commands (ls, grep, python3) and modify the agent's persistent memory via the add_memory tool call.
Sanitization: There are no sanitization or filtering steps described to validate the content being imported into the memory system, allowing instructions embedded in those files to influence future agent behavior.

import