The Agent Skills Directory

[SAFE]: The skill follows standard operational procedures for AI agent task management, including clear planning and approval steps.\n- [COMMAND_EXECUTION]: The skill utilizes the bash tool and specific CLI utilities such as 1c-syntax-checker and 1c-help for code validation and assistance. These tools are appropriate for its stated purpose of development and project implementation.\n- [PROMPT_INJECTION]: The instructions are clear and focused on task routing and design. There are no attempts to bypass safety filters, extract system prompts, or override agent constraints.\n- [EXTERNAL_DOWNLOADS]: No remote code execution or external downloads from untrusted sources were identified. Resources mentioned, such as 1c-syntax-checker, are specialized development tools consistent with the skill's domain.\n- [DATA_EXFILTRATION]: No patterns of sensitive file access or network-based data exfiltration were detected.

brainstorm