arguments-test-skill_skill

Overview

This skill tests argument substitution behavior for agent skills. It verifies how multiple placeholders are replaced, how escaped placeholders remain literal, and how absence of trailing arguments is handled. Use it to validate prompt templating and runtime argument injection.

How this skill works

The skill accepts a string of input that may include the placeholder $ARGUMENTS in multiple locations. During execution the skill replaces each $ARGUMENTS occurrence with the provided argument payload while leaving escaped $$ARGUMENTS unchanged as a literal. It also handles cases where no arguments are supplied so that templates with no trailing arguments do not cause errors.

When to use it

• Validating prompt templating logic and placeholder expansion.
• Testing agents that inject runtime arguments into skills or prompts.
• Confirming escape behavior for literal placeholder text ($$ARGUMENTS).
• Checking consistent substitution when $ARGUMENTS appears multiple times.
• Ensuring safe handling when no arguments are provided.

Best practices

• Provide representative argument payloads to cover edge cases (empty, multiline, special characters).
• Include templates that repeat $ARGUMENTS to validate consistent substitution across multiple positions.
• Use escaped $$ARGUMENTS when you need the literal token to appear in outputs or logs.
• Run automated tests that assert both replaced and literal instances are handled correctly.
• Combine with integration tests to validate behavior inside end-to-end agent flows.

Example use cases

• Unit test that injects a single argument and asserts every $ARGUMENTS is replaced identically.
• Integration test where templates mix $ARGUMENTS and $$ARGUMENTS to verify literals remain intact.
• Regression test ensuring that missing arguments do not append or crash the pipeline.
• Prompt-engineering experiments to confirm how runtime values propagate into chained prompts and LLM calls.