agentscope-ai/OpenJudge
OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards
OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards
npx skills add agentscope-ai/OpenJudgeOpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards
Modular skill definitions for LuomiNest agents. Skills are composable capability units loaded at runtime, providing domain-specific knowledge, structured workflows, and prompt templates for RAG retrieval and memory management.
我的 Claude Code 自建 Skills 集合 — github-push / CV-create / llm-pipeline-scaffold / arxiv-paper-reporter
Modular research-agent skills: GPU jobs, claim verification. Extracted from a working autonomous-research lab.
Claude Code skills for the AlphaAI MCP — relevance-scored financial news, insider data, and market workflows inside your agent.
Long-term memory layer for Claude Code and Codex — markdown-in-git knowledge base, SQLite FTS5 retrieval, MCP, and native agent hooks.
Codex skill for paper review comments with HTML evidence pages.