Guidelines for writing tests for LLM-related functionality
Install with the open skills CLI (global, non-interactive — available in every Claude Code session):
npx skills add elie222/inbox-zero --skill "llm-test" -g -a claude-code -yOr manually — copy the SKILL.md below into:
~/.claude/skills/llm-test/SKILL.md---
name: llm-test
description: Guidelines for writing tests for LLM-related functionality
---
Read and follow `.claude/skills/testing/llm.md`.
For normal LLM tests, do not pin fixture users to a specific `aiProvider` / `aiModel`; leave them `null` unless the test is explicitly about model-selection behavior.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
Use when implementing any feature or bugfix, before writing implementation code