Wikiskill Grader — Data & AI agent for Claude Code
Strict LLM-judge grader — given a task's success criteria and an inference agent's transcript/final answer, returns a pass/fail verdict with a short reason.
How to install Wikiskill Grader
Installs to ~/.claude/agents/haoch-wikiskill-wikiskill-grader.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/haoch/WikiSkill/HEAD/agents/wikiskill-grader.md -o ~/.claude/agents/haoch-wikiskill-wikiskill-grader.md Restart Claude Code, or start a new session, for it to be picked up.
What Wikiskill Grader does
name: wikiskill-grader description: Strict LLM-judge grader — given a task's success criteria and an inference agent's transcript/final answer, returns a pass/fail verdict with a short reason. tools: Read
You are a strict, literal-minded grader. You will be given a task's prompt, its `success_criteria`, and an inference agent's transcript and final answer. Judge only whether the final answer satisfies the stated success criteria — do not reward good style, effort, or reasoning quality b
Alternatives in Data & AI
- Lemonade Specialist — Lemonade Server and SDK specialist for local LLM on AMD hardware 1.5k ★
- Rlm Subcall — Acts as the RLM sub-LLM (llm_query) 382 ★
- Gn Voice Judge — gn-voice 윤문 결과물의 문체 품질을 판정하는 심사 에이전트 93 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Video Footage Scout
Finds and extracts short, premium, rights-safe real footage for a video project, either from an internal LMS/r
Askit Quality Grader
Judges whether a skill triggers and behaves correctly by running it against its eval-set and grading the outpu
Eval Inline Instruction
Eval arm inline-instruction (Q4 — the CALCULATED vs MEASURED line reduced to its INSTRUCTION half: what to out
AI Coach Grader
Independent, strict grader for the student's AI Engineer course in ~/ai-engineer-course. Use only when the /ai
Scene Prompt Smith
Use this agent when drafting kie.ai Nano Banana 2 image prompts for the scenes of a short video. It turns a sc
Melchior
MELCHIOR-1, the scientist. Judges a proposition on logic, data, and correctness. Returns a strict JSON verdict
Related Skills
Ad Comment Moderation
Agent Skill for moderating comments under Facebook and Instagram ads: a rule engine that states a reason for e
Implement Plan
You are tasked with implementing an approved technical plan from thoughts/shared/plans/. These plans contain p
Create E2e
Create a Playwright E2E test with explicit success criteria