feat: add Atlas Cloud as an LLM provider

[binyangzhu000-sudo]

What & Why

Adds Atlas Cloud as a new LLM provider, in the same category as the existing OpenRouter provider. Atlas Cloud is an OpenAI-compatible gateway to 100+ LLMs from a single endpoint (https://api.atlascloud.ai/v1), identified by vendor/model ids (e.g. openai/gpt-4.1-mini, anthropic/claude-sonnet-4.6, deepseek-ai/deepseek-v3.2).

Disclosure: I work at Atlas Cloud.

Implementation notes

This repo has two directories that both carry per-provider files named e.g. openrouter.ts:

• web/src/config/ai-models/ (51 files) — a static model-metadata catalog. Nothing under web/src actually imports this directory.
• web/src/config/llm-providers/ — the real ModelProviderCard registry. DEFAULT_MODEL_PROVIDER_LIST is built from it, and llm-providers-adapter.ts's getDefaultBaseURL()/getProviderTemplate() (which read from this list) are what services/llm-chat.ts actually calls to build the real fetch() request to /chat/completions.

This PR only touches the second one, which is the one that matters for a working integration:

• web/src/config/llm-providers/atlascloud.ts (new): a ModelProviderCard mirroring groq.ts's shape (no special disableBrowserRequest, unlike OpenRouter), listing 9 models with real context-window values pulled from Atlas Cloud's own /v1/models catalog.
• web/src/config/llm-providers/index.ts: import + added to DEFAULT_MODEL_PROVIDER_LIST (next to OpenRouterProvider) + card re-export.
• web/src/config/llm-providers-adapter.ts: getDefaultBaseURL() and getDefaultCheckModel() each got one entry.

Testing

$ pnpm run build   # tsc && vite build
✓ built in 2.44s

$ pnpm run lint    # eslint . --ext ts,tsx --max-warnings 0
(no errors)

(This project has no unit test suite.)

Manual test — a real call to the exact endpoint this config resolves to:

$ curl -X POST https://api.atlascloud.ai/v1/chat/completions \
    -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \
    -d '{"model":"openai/gpt-4.1-mini","messages":[{"role":"user","content":"Reply with exactly: UNLA_ATLASCLOUD_OK"}]}'
# {"choices":[{"message":{"content":"UNLA_ATLASCLOUD_OK", ...

🤖 Generated with Claude Code

Summary by Sourcery

Add Atlas Cloud to the available LLM providers and expose its supported models through the provider registry.

New Features:

• Add Atlas Cloud as an OpenAI-compatible LLM provider with access to nine configured models.

Enhancements:

• Register Atlas Cloud’s API endpoint and default validation model for request construction and provider checks.

[sourcery-ai]

Reviewer's Guide

This PR integrates Atlas Cloud as a first-class LLM provider by adding an OpenAI-compatible provider card, registering nine catalog models, and wiring its API endpoint and default check model into the provider adapter; build and lint reportedly pass, with an additional successful manual chat-completion request.

Sequence diagram for an Atlas Cloud chat completion

sequenceDiagram
    participant Chat as Chat service
    participant Adapter as LLM provider adapter
    participant Atlas as Atlas Cloud API

    Chat->>Adapter: getDefaultBaseURL(atlascloud)
    Adapter-->>Chat: https://api.atlascloud.ai/v1
    Chat->>Atlas: POST /chat/completions with model openai/gpt-4.1-mini
    Atlas-->>Chat: Chat completion response

Loading

File-Level Changes

Change	Details	Files
Add Atlas Cloud as an OpenAI-compatible provider with its endpoint, default validation model, and supported model catalog.	Define nine enabled chat models with provider-specific IDs, context windows, function-calling support, and vision capability where applicable. Configure the Atlas Cloud base URL and default check model used for request construction and provider validation. Register and re-export the provider card in the default provider registry.	web/src/config/llm-providers/atlascloud.ts web/src/config/llm-providers-adapter.ts web/src/config/llm-providers/index.ts

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey - I've found 1 issue

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="web/src/config/llm-providers/atlascloud.ts" line_range="85" />
<code_context>
+    'Atlas Cloud 是一个 OpenAI 兼容的统一网关，通过单一 API 接入 100+ 大语言模型，涵盖 OpenAI、Anthropic、DeepSeek、Qwen、Z.AI、Moonshot、MiniMax 等厂商，模型 ID 采用 `vendor/model` 格式。',
+  id: 'atlascloud',
+  modelList: { showModelFetcher: true },
+  modelsUrl: 'https://www.atlascloud.ai/',
+  name: 'Atlas Cloud',
+  proxyUrl: {
</code_context>
<issue_to_address>
**nitpick:** The Atlas Cloud provider's model documentation link opens the provider homepage rather than a model catalog or model documentation page, so the UI's “view models” link does not take users to the advertised model information.

**Suggested fix:** Point `modelsUrl` to Atlas Cloud's actual model catalog or API documentation URL.

```suggestion
  modelsUrl: 'https://www.atlascloud.ai/models',
```
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

[binyangzhu000-sudo]

Thanks for the review, Sourcery. Verified against the current code: modelsUrl pointed at the Atlas Cloud homepage, so the "view models" link didn't land on a catalog. Fixed in the latest commit by pointing it to https://www.atlascloud.ai/models (this URL returned HTTP 200 when I checked). This is a one-line string change; I did not re-run the full build/lint locally for it.