The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Connects to official Linear MCP server endpoints at https://mcp.linear.app/mcp and https://mcp.linear.app/sse for service integration as specified in SKILL.md and agents/openai.yaml.- [COMMAND_EXECUTION]: Outlines standard configuration steps including codex mcp add, codex --enable, and codex mcp login for local setup.- [COMMAND_EXECUTION]: Provides a Windows/WSL configuration using npx to run the official mcp-remote tool from Linear's infrastructure.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from an external project management workspace.
Ingestion points: Processes data from issues, comments, and documentation via get_issue, list_comments, and get_document in SKILL.md.
Boundary markers: Absent. No explicit markers or safety instructions are defined to separate ingested workspace content from core agent instructions.
Capability inventory: The skill allows modification of the workspace through tools like update_issue, create_comment, and update_project in SKILL.md.
Sanitization: Absent. No evidence of content validation or escaping is provided for processed data.

linear