The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the Sequential Thinking MCP server using standard package managers (npx) or container runtimes (Docker). These instructions point to official @modelcontextprotocol and mcp repositories, which are the standard sources for this protocol.
[COMMAND_EXECUTION]: The README contains configuration examples for claude_desktop_config.json that execute the npx or docker commands. These commands are necessary to launch the MCP server that provides the reasoning capabilities defined in the skill.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and sequence 'thoughts' as string inputs. While this constitutes an ingestion surface for potentially untrusted data, the skill is focused on internal reasoning logic and lacks high-privilege write capabilities or unsafe interpolation patterns that would enable a successful indirect injection attack.

sequential-thinking