From b3c177718c77608fe0c103c648f54eb5314ac22c Mon Sep 17 00:00:00 2001 From: Aaron Stannard Date: Sat, 1 Aug 2026 20:43:21 +0000 Subject: [PATCH] docs: add DeepSeek provider to managed-providers and architecture overview MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Closes #110 (docs: add DeepSeek provider after Netclaw 0.25.2 ships) - Added DeepSeek row to Provider Summary table in managed-providers.md - Added dedicated DeepSeek section following Venice.ai pattern - Updated architecture/overview.md LLM providers table (also backfilled GitHub Copilot and Venice.ai which were missing) - Updated Resources section with DeepSeek API docs link - Updated Configuration Schema Type field to include deepseek Screenshots (init-01-provider-list.png, provider-manager.png) need regeneration after 0.25.2 ships — the provider picker and TUI won't show DeepSeek until the binary supports it. --- src/content/docs/architecture/overview.md | 5 ++++- .../docs/configuration/managed-providers.md | 20 ++++++++++++++++--- 2 files changed, 21 insertions(+), 4 deletions(-) diff --git a/src/content/docs/architecture/overview.md b/src/content/docs/architecture/overview.md index 27c1018..1ae13f0 100644 --- a/src/content/docs/architecture/overview.md +++ b/src/content/docs/architecture/overview.md @@ -98,7 +98,10 @@ Netclaw uses [Microsoft.Extensions.AI](https://learn.microsoft.com/en-us/dotnet/ | **Ollama** | `ollama` | Code default. Native integration, queries `/api/show` for capabilities. | | **OpenRouter** | `openrouter` | Recommended for cloud models. Widest model catalog. | | **Anthropic** | `anthropic` | Direct API. | -| **OpenAI** | `openai` | Direct API. | +| **OpenAI** | `openai` | Direct API, OAuth or API key. | +| **GitHub Copilot** | `github-copilot` | OAuth device flow. Uses your Copilot subscription. | +| **Venice.ai** | `veniceai` | Privacy-focused, OpenAI-compatible API. | +| **DeepSeek** | `deepseek` | 1M context window, thinking mode, text-only. | | **OpenAI-compatible** | `openai-compatible` | Generic self-hosted endpoint. | At startup, the daemon queries the provider to detect context window size, vision support, and tool use capabilities. If the direct query fails, it falls back to the OpenRouter catalog, then HuggingFace metadata, then assumes text-only. diff --git a/src/content/docs/configuration/managed-providers.md b/src/content/docs/configuration/managed-providers.md index c42a245..017af17 100644 --- a/src/content/docs/configuration/managed-providers.md +++ b/src/content/docs/configuration/managed-providers.md @@ -1,9 +1,9 @@ --- title: Managed Providers -description: Configure cloud-hosted LLM providers — OpenRouter, Anthropic, OpenAI, GitHub Copilot, and Venice.ai. +description: Configure cloud-hosted LLM providers — OpenRouter, Anthropic, OpenAI, GitHub Copilot, Venice.ai, and DeepSeek. --- -Managed providers are cloud-hosted LLM services that netclaw connects to over HTTPS. Each authenticates with an API key or OAuth: OpenRouter, Anthropic, OpenAI, GitHub Copilot, and Venice.ai. +Managed providers are cloud-hosted LLM services that netclaw connects to over HTTPS. Each authenticates with an API key or OAuth: OpenRouter, Anthropic, OpenAI, GitHub Copilot, Venice.ai, and DeepSeek. Provider config lives in two files. Non-secret fields (type, endpoint, auth method) go in `~/.netclaw/config/netclaw.json`. Credentials go in `~/.netclaw/config/secrets.json`, which is [encrypted at rest](/security/secrets/) — write plaintext values and netclaw encrypts them on first read. Environment variables override both. @@ -20,6 +20,7 @@ For self-hosted inference (Ollama, llama.cpp, vLLM), see [Self-Hosted Providers] | `openai` | OpenAI | `https://api.openai.com` | OAuth or API key | [platform.openai.com](https://platform.openai.com/api-keys) | | `github-copilot` | GitHub Copilot | `https://api.githubcopilot.com` | OAuth (device) | GitHub Copilot subscription | | `veniceai` | Venice.ai | `https://api.venice.ai/api/v1` | API key | [venice.ai/settings/api](https://venice.ai/settings/api) | +| `deepseek` | DeepSeek | `https://api.deepseek.com/v1` | API key | [platform.deepseek.com/api_keys](https://platform.deepseek.com/api_keys) | ## Configuration Schema @@ -27,7 +28,7 @@ Each provider is a named entry under the `Providers` section. The key is a name | Field | Type | Default | Description | |-------|------|---------|-------------| -| `Type` | string | `"ollama"` | Provider SDK: `openrouter`, `anthropic`, `openai`, `github-copilot`, or `veniceai` for managed providers. Always set this explicitly. | +| `Type` | string | `"ollama"` | Provider SDK: `openrouter`, `anthropic`, `openai`, `github-copilot`, `veniceai`, or `deepseek` for managed providers. Always set this explicitly. | | `Endpoint` | string | `""` | Base URL. Leave empty to use the provider's default | | `ApiKey` | string? | `null` | API key. Store in `secrets.json`, never `netclaw.json` | | `AuthMethod` | enum | `None` | `None`, `ApiKey`, `OAuthDevice`, or `OAuthPkce` | @@ -236,6 +237,18 @@ By default, netclaw tells Venice **not** to prepend its own system prompt (`incl } ``` +## DeepSeek + +[DeepSeek](https://api-docs.deepseek.com/) is a hosted inference provider with a 1M-token context window, tool calls, JSON output, and optional thinking mode. Auth is an API key: + +```bash +netclaw provider add deepseek deepseek --api-key sk-... +``` + +The endpoint defaults to `https://api.deepseek.com/v1`. Get a key at [platform.deepseek.com/api_keys](https://platform.deepseek.com/api_keys). + +Netclaw supports two hosted DeepSeek models: `deepseek-v4-pro` and `deepseek-v4-flash`. Both are text-only — no vision, no audio. The 1M context window and modality constraints are applied automatically since DeepSeek's `/models` endpoint doesn't report capabilities for the V4 family. + ## Provider Manager TUI ![Provider Manager TUI showing configured providers with health status](/screenshots/output/provider-manager.png) @@ -310,5 +323,6 @@ To test your configuration right now, run [`netclaw doctor`](/cli/doctor/) for a - [OpenAI API docs](https://platform.openai.com/docs/api-reference) — endpoints, authentication, model capabilities - [GitHub Copilot](https://github.com/features/copilot) — subscription details and supported models - [Venice.ai API docs](https://docs.venice.ai/) — endpoints, models, and vendor parameters +- [DeepSeek API docs](https://api-docs.deepseek.com/) — authentication, models, and rate limits - [.NET environment variable configuration](https://learn.microsoft.com/en-us/dotnet/core/extensions/configuration-providers#environment-variable-configuration-provider) — the double-underscore nesting convention - [RFC 7636 — OAuth PKCE](https://datatracker.ietf.org/doc/html/rfc7636) — the code exchange flow OpenAI uses