研究库 开源仓库
Repositories · organized/repo_cards

仓库/Skill 库

4 个 · 评测基准 · 模型

排序 Stars 周增
bespokelabsai/nimble
Python · 2026-09-23 评测基准 模型 研究原型 Stars 1647 周增 +0

本地类型化决策、对比式数据策展与模型评估。Local typed decisions, contrastive data curation, and model evaluation.

evaluation
PerryLink/dsh-auto-review
TypeScript · 2026-10-05 评测基准 模型 实验 Stars 233 周增 +0

DeepSeek Harness 审批请求的 second-model AI 自动复核:只读复核子 Agent 返回结构化 allow/deny 判定及理由,默认 fail-closed,会话日志(approval/asked → autoReview/verdict → approval/decided)全程可审计。Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default, fully auditable from the session log (approval/asked -> autoReview/verdict -> approval/decided).

agentriskllm-infra
Totoro-qaq/dsh-plugin-bridge
JavaScript · 2026-08-22 评测基准 模型 实验 Stars 77 周增 +0

DeepSeek Harness 插件,支持跨预设会话迁移的可预览;固定 schema 的交接保留状态、源模型意图与未处理图像,原始会话保持不变DeepSeek Harness plugin for previewable cross-preset session migration. Fixed-schema handoffs preserve state, source-model intent, and unresolved images; the original session stays untouched.

multimodal
sb1733831438-maker/DSH-closerAI
TypeScript · 2026-08-18 评测基准 模型 实验 Stars 4 周增 +0

CloserAI — 基于 DeepSeek Harness 构建的本地优先、模型无关、权限透明的桌面 AI 工作台。CloserAI - a local-first, model-agnostic, permission-transparent desktop AI workbench built on DeepSeek Harness.

agent