免费大模型一键接入（free-llm-setup）：把免费额度接进 WorkBuddy 当自定义模型用
介绍社区 Skill free-llm-setup：一键把免费大模型接进 WorkBuddy 当自定义模型，这部分对话不消耗平台积分。主推「一个 Key 通吃」——1 个魔搭 Token 换 48 个模型、每天 2000 次免费、次日零点重置；境外模型再加 OpenRouter（免费池全量实测）。共支持 21 家平台（4 家聚合池 + 17 家单家）。能力包括自动探测代理、对每个服务做通道竞速、区分 Cloudflare 403 拦截与真的不通、实测通过才写入、按延迟排序、写入前备份、自动清理下架条目。附带体检脚本（体检已接入模型/删死条目/免费池补员/出 HTML 报告），按需手动跑，每日自动化已于 2026-10-08 撤除。适合想省积分、没有会员、积分用完或被限流、要接自定义模型、国外模型挂了代理仍不流畅的场景。
Read Moredsh-custom-provider: Add Passwordless Authentication Toggle for Custom Model Providers
The DeepSeek Harness (DSH) plugin “dsh-custom-provider” allows local or self-hosted model endpoints to bypass API key validation. It adds a “Passwordless Authentication” toggle to manually declared model provider cards and uses a green status dot to reflect authentication status. Users can enable this option in settings to resolve interruptions caused by endpoints that do not require authentication. The plugin indicates status by checking whether the `Authorization` header is non-empty, making it suitable for integrating local LLM services.
Read MoreDeepSeek Harness (DSH) Custom LLM Provider Plugin
`dsh-custom-provider` is a plugin for DeepSeek Harness (DSH) designed to address issues such as inconsistent protocols and difficulties in dynamically retrieving model lists when integrating custom LLM services. The plugin supports multiple protocols, including OpenAI and Anthropic, through static configuration, and reuses the DSH streaming adapter. Installation requires running a command in the Web profile, and it supports both the Web UI and declarative configuration via `settings.yaml`. The plugin features visual management, static model catalogs, per-request credential resolution, and compatibility field overrides. Before use, ensure that the Node.js version and DSH version are compatible, and that the configured `baseURL` conforms to the expectations of the selected protocol adapter.
Read MoreAdd Visual, Reasoning Tier and Token Capacity for Custom Models in DeepSeek Harness via better-model-provider
### 正式翻译： The official "Models" page of DeepSeek Harness can manage providers and API keys, but in the past, adjusting the inference strength levels, visual input capabilities, and token capacity for custom models could only be done by manually editing the YAML files. The MIT-licensed plugin `better-model-provider` maintained by sanshanya added a "Model Capabilities" section on the settings page, which configures these native capabilities based on the model's declared specifications, without modifying the Harness runtime or touching credentials. This article verifies the installation commands, usage methods, and compatibility boundaries based on the community directory page and the GitHub README.
Read More