Spell Tester — Development agent for Claude Code
Subagent dispatched during the try-it phase of building-a-spell.
How to install Spell Tester
Installs to ~/.claude/agents/redhuntlabs-wizard-spell-tester.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/redhuntlabs/wizard/HEAD/agents/spell-tester.md -o ~/.claude/agents/redhuntlabs-wizard-spell-tester.md Restart Claude Code, or start a new session, for it to be picked up.
What Spell Tester does
name: spell-tester description: Subagent dispatched during the try-it phase of building-a-spell. Tests a draft spell against realistic scenarios and returns a verdict. Has TWO MODES based on the draft's kind. tools: Read, Write, Bash
Spell Tester
You have been dispatched as a subagent to test a draft spell. Your job is to attempt the draft on realistic scenarios and return one of three verdicts:
- PASS — the draft works as written, no changes needed
- NEEDS-REFINEMENT — the
Alternatives in Development
- TDD Refactoring Agent — Improves code structure while keeping all tests green during the TDD REFACTOR phase using proven refactoring p 652 ★
- Executioner — Optional Sanhedrin fallback verifier 632 ★
- API Tester Specialist.Agent — Specialist in creating and executing API tests 225 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Skill Tester
Tests skill functionality by simulating usage scenarios and validating outputs
Fiori Tester
INTERNAL subagent of /sap-fiori — never invoke directly. Only called by the Fiori parent agent during the test
Prose Judge
Fresh-context fidelity judge for the prose bench. Reads ONE gate file (docs/prose-rubric.md, the materials, on
Issue Court
Try one open issue of this repository's machine population per run under a short adversarial review, and post
Business Idea Critic
Read all ten rubric results for one business idea and try to refute every score of 1, 2 or 5 and every confide
Weavie Tester
Proves a change actually works by running the real Weavie app, exercising the scenarios a PR should cover, and
Related Skills
Helpful Content Review Skill
A Claude skill that reviews and scores any draft or live page against Google's helpful, reliable, people-first
Author Council
Convene a pluggable author council — routes to sub-rosters (fiction / magic-system / worldbuilding / prose / m
AI Tester
End-to-end behavioral testing skills, bare system prompts, and any agent runtime — run real scenarios in an is