The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides the execute_terminal_command tool, allowing the agent to perform operations in the IDE's integrated terminal. Documentation notes that shell features are disabled by default unless explicitly requested.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted project data (source code, symbol definitions, and compilation errors) that could contain malicious instructions intended to manipulate the agent's behavior.
Ingestion points: Found in search_text, search_symbol, and get_file_problems tools which return code content to the agent.
Capability inventory: The skill can execute terminal commands, rename symbols globally, and replace text in files.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat code content as data rather than instructions.
Sanitization: No sanitization or validation of the processed code content is documented.

idea-mcp