The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands such as npx tsx, node, and the linear CLI to perform operations on the Linear platform (detailed-guide.md).
[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes external data from the Linear platform, such as issue descriptions and comments, which could contain malicious instructions.
Ingestion points: Linear issues, project data, and comments fetched via the API or CLI (detailed-guide.md).
Boundary markers: No explicit delimiters or guardrail instructions are provided to the agent to treat fetched data as untrusted (SKILL.md, detailed-guide.md).
Capability inventory: The skill possesses the capability to execute commands, perform network requests, and modify external platform data (detailed-guide.md).
Sanitization: The instructions do not specify any validation or sanitization steps for data retrieved from Linear.
[DYNAMIC_EXECUTION]: Employs node -e to execute JavaScript code snippets directly from the terminal for environment validation (detailed-guide.md).
[PERSISTENCE]: Provides instructions for the user to manually append environment variable configurations to shell startup files like ~/.zshrc and ~/.bashrc (detailed-guide.md).
[EXTERNAL_DOWNLOADS]: Configures an MCP server to connect to an external endpoint (https://mcp.linear.app/mcp), facilitating network communication with the Linear platform.

linear-claude-skill