dsh-llm-friendli
Run the following command in DeepSeek Harness:
dsh plugin install Lee-Si-Yoon/dsh-llm-friendli
Paste the following prompt into your AI chat to install this plugin:
Run the dsh plugin install command to add this plugin to your DeepSeek Harness profile; the full source is available at https://github.com/Lee-Si-Yoon/dsh-llm-friendli
About this plugin
DeepSeek Harness ships with a limited set of built-in inference providers, and FriendliAI models simply cannot be plugged in. dsh-llm-friendli bridges that gap by implementing the harness LlmAdapter contract on top of Friendlis OpenAI-compatible chat API, registering Friendli as a first-class provider with zero configuration required.
Two capabilities stand out. First, the model catalog is fetched live from Friendli at runtime, so newly available models appear in the dsh web UI immediately without any manual ID management. Second, the API key is always read from the environment on every request, keeping secrets out of config files and version control by design.
This plugin is for anyone who already has a FriendliAI account and wants to pick a Friendli model directly inside the dsh web interface for chat or inference, as well as developers who need to point an external agent at a specific Friendli model ID through the Harness pipeline.
Use Cases
- Pick a Friendli model directly in the dsh web UI for chat or inference
- Point an external agent at a specific Friendli model through the Harness pipeline
- Auto-discover newly available Friendli models without manual ID management
Best For
- Developers with a FriendliAI account looking to integrate into the Harness ecosystem
- LLM application builders who need to switch between multiple inference providers
- Users who prioritize API key security and prefer environment variables over hardcoded config
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.