hello-world

name	hello-world
description	Minimal valid dual-runtime skill that prints a greeting in either Hermes or OpenClaw. Use when verifying a freshly-installed claw-hermes setup or when demonstrating the agentskills.io v1.0 runtimes extension to a contributor.
version	0.1.0
license	MIT
author	claw-hermes contributors
agentskills_version	1.0
runtimes	{"hermes":{"entrypoint":"python -c \"print('hello from hermes')\"","capabilities":[]},"openclaw":{"entrypoint":"node -e \"console.log('hello from openclaw')\"","capabilities":[]},"both":{"requires_capabilities":[]}}
keywords	["example","hello-world","dual-runtime"]
homepage	https://github.com/bertbertov/claw-hermes

hello-world

The smallest possible dual-runtime skill. Each runtime entrypoint prints a single line and exits. This file is intentionally short on prose because its purpose is to demonstrate the shape of a valid manifest, not to ship behavior. Use it as a template when you start a new skill, or feed it to claw-hermes skill lint as a sanity check that your local installation parses the v1.0 manifest schema correctly.

When to invoke this skill

• "verify my claw-hermes install is wired up"
• "show me a minimal dual-runtime SKILL.md"
• "what's the simplest manifest that lints clean?"

Inputs and outputs

There are no inputs. Each runtime entrypoint prints a fixed greeting to stdout and exits with code 0. The skill exists purely to exercise the manifest pipeline end-to-end without invoking any external services or capabilities.