DSH / BUNDLE / BUNDLES

treg-dsh

v0.27.0superdesigndev / tregaccad8a3ea

InstallableBundlesBundles & other modulesCommunity · Topic auto-analysis

Open on GitHub

Overview Installing Source Related by capability

Overview

treg-dsh

OpenRouter for tools - 4,000+ agent-friendly tools, pay for the usage, not subscription

README / EN

Package documentation

Registry summary

OpenRouter for tools - 4,000+ agent-friendly tools, pay for the usage, not subscription

dsh.pub verifies the pinned bundle contract, runtime facts, and distribution semantics. The complete README remains in the source repository.

Read the full README on GitHub

Install

Install from GitHub

The CLI validates the bundle, invokes native DSH, then reports completion.

$npx dshpub add superdesigndev/treg --ref accad8a3eac6b09b439e62a85480a636b0b90588

—CLI-reported installs

Completed installs reported by the dshpub CLI. This is not unique users, repository downloads, or native DSH installs.

Source

Version: 0.27.0
License: Apache-2.0
Type: Bundles
Capability: Bundles
Source scope: Community · Topic auto-analysis
Commit: accad8a3ea
Path: .

Open on GitHub

Same capability topic

Related by capability

Ranked by shared capability topic, runtime face, tools or UI contributions, and distribution mode. This is not a similarity, quality, or compatibility ranking.

@towzai/dsh-memoryDSH 跨会话记忆插件 | Cross-session memory plugin: YAML 存储、embedding 语义检索、标题优先注入、两级自动注入Bundles & other modules · Installable
trae-cn-providerTrae CN as a native DeepSeek Harness LLM provider: browser OAuth login by default, optional local-install credential source, OpenAI-compatible surface on the host web serverBundles & other modules · Installable
@xiciyang/training-pretrain继续预训练（continued pretraining）薄插件：领域语料校验（双轨去泄漏 + 全量 PII）→ 公共训练循环（training-core，容器内 transformers）→ 分片续训（步号与数据位置连续性机器可验）→ 合并 → serving 部署 → 遗忘门禁。依赖长任务 1 冻结的六个 @ai-infra/* 共享库。Bundles & other modules · Installable
@xiciyang/training-qatQAT 薄插件：语料 O1 去泄漏 → 公共单体训练循环（training-core，容器内 torchao QAT 假量化）→ 对 QAT 训练后的权重施加真实的 INT8 量化（仓库自己的 llmcompressor INT8，compressed-tensors 产出）→ serving 部署 + 双门禁 + 跨插件对账。依赖长任务 1 冻结的六个 @ai-infra/* 共享库。Bundles & other modules · Installable
@xiciyang/training-rlRL（GRPO）薄插件：规则奖励（显式版本）→ β 可推导（β=σ_r/KL*，σ_r 实测）→ training-core 单体档位驱动 TRL GRPOTrainer → 产物级断言 → 合并 → serving 部署 → 偏好对齐 + 遗忘门禁。依赖长任务 1 冻结的六个 @ai-infra/* 共享库。Bundles & other modules · Installable
@xiciyang/training-sftSFT (LoRA) 薄插件：数据校验 → 公共训练循环（training-core，容器内 peft） → 合并 → serving 部署 → 评估门禁。依赖长任务 1 冻结的六个 @ai-infra/* 共享库。Bundles & other modules · Installable

Bundles & other modules How installable plugins work Registry concepts