The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and connects to official Linear infrastructure at https://mcp.linear.app/mcp and https://mcp.linear.app/sse for its Model Context Protocol (MCP) server functionality.
[COMMAND_EXECUTION]: Instructions are provided for the user to configure the environment using the codex CLI and npx. These include adding the MCP server, enabling the remote client, and performing OAuth login to secure access to the Linear workspace.
[DATA_EXFILTRATION]: The skill processes content from the user's Linear workspace, which involves inherent ingestion of external data.
Ingestion points: Issue details, search results, and comments fetched via tools such as get_issue and search_documentation in SKILL.md.
Boundary markers: None identified; the agent relies on its internal safety guidelines when processing fetched content.
Capability inventory: The skill can perform write operations back to Linear, such as create_issue, update_issue, and create_comment, as documented in SKILL.md.
Sanitization: No explicit sanitization logic is present within the skill's instructions for the fetched data.
[SAFE]: No malicious behavior, obfuscation, or unauthorized data exfiltration patterns were detected. The skill's operations are consistent with its stated purpose of managing a Linear project management workflow.

linear