hccccc01333/dsh-eval · 18 天前更新
安装方式
dsh plugin --profile eval add dsh-evaldsh plugin --profile web add github:hccccc01333/dsh-evalAgent evaluation platform: benchmark YAML, headless run orchestration, trace-based metrics, and run reports
@deepseek-ai/cordis^4.0.1共享@deepseek-ai/dsh-cmdline^0.1.0-rc.6共享@deepseek-ai/dsh-invariants^0.1.0-rc.6共享@deepseek-ai/dsh-llm^0.1.0-rc.6共享@deepseek-ai/dsh-llm-retry^0.1.0-rc.6共享@deepseek-ai/dsh-session^0.1.0-rc.6共享路径固定于提交 47f39d7c1453
Agent evaluation platform for DeepSeek Harness: benchmark YAML, headless dsh orchestration, trace-based metrics, LLM judge, paired A/B, keyless replay, and cross-harness import.
结果绑定到所列 commit，仅涵盖本站实际执行的静态检查。证据分衡量可复核信息的完整度，不是兼容性百分比，也不构成安全保证。
47f39d7c1453兼容性置信度: 有限
packages/eval/package.json#prepackpnpm run buildinstall.scriptpackages/eval/package.jsoninstall-time"prepack": "pnpm run build"扫描器 1.0.0 · 评分器 1.0.0