Eval Author — Testing & QA agent for Claude Code
Authoring role for the D365FO agent eval loop catalog.
How to install Eval Author
Installs to ~/.claude/agents/dynamics365ninja-d365fo-mcp-server-eval-author.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/dynamics365ninja/d365fo-mcp-server/HEAD/.claude/agents/eval-author.md -o ~/.claude/agents/dynamics365ninja-d365fo-mcp-server-eval-author.md Restart Claude Code, or start a new session, for it to be picked up.
What Eval Author does
name: eval-author description: Authoring role for the D365FO agent eval loop catalog. Drafts a new eval/cases/.json spec (valid against eval/cases/schema.json), scaffolds its golden folder, and sets golden_pending until the golden is captured on the VM. Use when asked to "add an eval case", "author a case for ", "draft a case from this failure", or write a new use-case for the catalog. tools: Bash, Read, Edit, Write, Grep, Glob model: inherit
You author new cases for the ev
Alternatives in Testing & QA
- Prod Incident Responder — Reads recent prod logs, picks the highest-impact recurring error, drafts a minimal fix PR with a regression te 324 ★
- Unit Executor — Implements exactly one unit of work from an inline spec and returns a structured JSON record 314 ★
- Shep Use Case Creator — Scaffolds a new application-layer use case with TDD (RED test file first, then minimal GREEN implementation) 247 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Eval Improver
Improver role of the D365FO agent eval loop. Reads the corpus of run records, ranks failure clusters, reproduc
Osf Clean Room
Reads a feature inside a temp folder, extracts its behavior and test coverage as a source-free specification,
AI Eval Designer
Use this agent to design a risk-tiered evaluation set for an AI feature. Trigger when the user says "design ev
Test Case Design
两阶段。阶段1=解构:读需求 + UI设计图 + page_map + 知识库,产出 decomp.md(7 张结构化表,GATE 0 人审)。阶段2=生成:读 confirmed decomp.md 查表映射出 cas
Rondoflow i18n Translator
Translates RondoFlow UI strings into Slovak (sk) and Spanish (es) and writes them into the locale catalogs, ke
Test Fixtures For Consistent Test Data
@pytest.fixture def valid_user(): return User( id="test-123", email="test@example.com", name="Test User", crea
Related Skills
Claude Evals
Production eval framework for Claude Agent SDK — implements Anthropic's published eval patterns with native SD
Eval Cases
Grow an eval case set from the placeholder rows to a real 15-25 - harvest genuine cases out of the repository
Scaffold Plugin Test
Scaffold a headless behavior test for a werkstoff plugin skill — a seeded-defect fixture plus a case in test/p