The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from Linear issues, projects, and comments which are external and could contain malicious instructions.\n
Ingestion points: Data is retrieved through the linear CLI and GraphQL queries executed by local scripts (SKILL.md).\n
Boundary markers: There are no specified delimiters or instructions to treat external data as untrusted text rather than commands.\n
Capability inventory: The skill has the ability to execute shell commands via the linear CLI and run TypeScript scripts to modify project data (SKILL.md).\n
Sanitization: No sanitization logic is described for handling external Linear content before it enters the agent's context.\n- [DYNAMIC_EXECUTION]: Core functionality is implemented through local TypeScript scripts executed at runtime.\n
Evidence: Frequent use of npx tsx to run scripts such as scripts/setup.ts, scripts/query.ts, and scripts/linear-ops.ts (SKILL.md).\n- [REMOTE_CODE_EXECUTION]: The skill is designed to interact with the official Linear MCP server.\n
Evidence: Configuration instructions suggest using npx mcp-remote https://mcp.linear.app/sse to connect to Linear's official remote server (SKILL.md).\n- [COMMAND_EXECUTION]: The skill relies on executing the linear CLI for many of its operations.\n
Evidence: Documentation includes examples of linear issues view, linear issues create, and linear issues comment (SKILL.md).\n- [EXTERNAL_DOWNLOADS]: The skill uses npx to download and run the tsx and mcp-remote packages.\n
Evidence: Usage of npx tsx and npx mcp-remote implies downloading these tools from the NPM registry if they are not already cached (SKILL.md).

linear-claude-skill