Skip to content

feat: add Atlas Cloud provider for the standalone brain - #173

Open
binyangzhu000-sudo wants to merge 1 commit into
awarexone:mainfrom
binyangzhu000-sudo:feat/atlascloud-provider
Open

binyangzhu000-sudo wants to merge 1 commit into
awarexone:mainfrom
binyangzhu000-sudo:feat/atlascloud-provider

Conversation

@binyangzhu000-sudo

Copy link
Copy Markdown

Summary

Adds Atlas Cloud (https://api.atlascloud.ai/v1) as a provider for the standalone bughunter / brain.py, following the Requesty addition (#159). It is an OpenAI-compatible multi-model gateway.

Disclosure: I work at Atlas Cloud.

What changed

  • bughunter/brain.py: atlascloud in PROVIDER_KEY_ENV (ATLASCLOUD_API_KEY), DEFAULT_MODELS (deepseek-ai/deepseek-v4-flash), _init_provider (bearer session, base URL from ATLASCLOUD_BASE_URL or the default), the OpenAI-compatible chat branch, and a curated list_models(). Like Requesty it is not in PROVIDER_PRIORITY: it is only picked when selected explicitly or when ATLASCLOUD_API_KEY is set, so it never preempts a provider you already configured.
  • bughunter/engine.py: option 12 in bughunter setup, the key map in setup and providers, the --provider choices and help, the config-to-env key loop.
  • README.md, CHANGELOG.md, config.example.json: provider table row, opt-in note, changelog entry, example config line.
  • tests/test_brain_auto_detect.py: 5 tests mirroring the Requesty ones (key moves it to the front, init and api base, base-URL override, no key means unavailable, opt-in only).

Test plan

  • pytest tests (excluding tests/benchmark): 1127 passed; main without this change gives 1122, so the 5 new tests account for the difference and nothing else changed.
  • Live check against the real Atlas Cloud API through LLMClient(provider="atlascloud") with only ATLASCLOUD_API_KEY set: it resolved https://api.atlascloud.ai/v1, and chat() returned the expected text for the default model and for moonshotai/kimi-k3, zai-org/glm-5.2 and qwen/qwen3.5-plus.
  • Every model id in the curated list exists in Atlas Cloud's /v1/models.

🤖 Generated with Claude Code

Atlas Cloud (https://api.atlascloud.ai/v1) is an OpenAI-compatible
multi-model gateway. Same shape as the Requesty addition: ATLASCLOUD_API_KEY
(optional ATLASCLOUD_BASE_URL), opt-in in auto-detect (a set key moves it to
the front, but it is not in PROVIDER_PRIORITY), option 12 in `bughunter
setup`, `--provider atlascloud`, a default model
(deepseek-ai/deepseek-v4-flash) and a short curated list, plus README,
CHANGELOG and config.example.json entries and tests.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Signed-off-by: binyangzhu000-sudo <224954946+binyangzhu000-sudo@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant