Nm Leyline Service Registry

Service registry patterns for managing external services, health checks, centralized configuration, and unified execution

This skill is an instruction-only pattern for managing and executing external service CLIs. Before installing: (1) understand that at runtime the agent will build and run arbitrary commands (e.g., gemini, qwen) and may read environment variables and local config files — review any service configs you add to avoid exposing secrets; (2) note the skill does not declare the API key env vars it expects (e.g., GEMINI_API_KEY), so double-check what environment variables will be accessed and avoid placing high-value credentials in plain env if you don't trust the skill; (3) inspect or restrict the ~/.claude/leyline/services.yaml and any health_check_cmd/auth_check_cmd values to ensure they do not execute unsafe commands; (4) run the skill with least privilege (not as root) and limit which services/config entries are registered; (5) if you need higher assurance, request the author to declare required env vars and explicit config paths in metadata or provide a trimmed/sandboxed implementation.

Type: OpenClaw Skill Name: nm-leyline-service-registry Version: 1.0.0 The skill provides a framework for managing and executing external CLI services via shell commands, which is an inherently high-risk capability. While it uses `shlex.split` in `modules/execution-patterns.md` to mitigate shell injection, the `SKILL.md` file contains highly repetitive instructions forcing the AI agent to execute shell commands for 'verification' purposes, which could be used to probe the host environment. Furthermore, the `_meta.json` file contains a future publication date (April 2026), which is often an indicator of automated or placeholder content generation.