deepseek-ai/awesome-deepseek-agent
1
0
Fork 0
mirror of https://github.com/deepseek-ai/awesome-deepseek-agent.git synced 2026-11-05 09:17:39 +00:00
Code Issues 354 Projects Packages Wiki Activity

[GH-ISSUE #465] Add integration guide: CoreCoder #134

New issue
Open
opened 2026-10-26 09:45:51 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-26 09:45:51 +00:00
Owner
Copy link

Originally created by @he-yufeng on GitHub (Oct 19, 2026).
Original GitHub issue: https://github.com/deepseek-ai/awesome-deepseek-agent/issues/465

I'd like to add an integration guide for CoreCoder, a minimal, fully readable coding-agent engine (~1.3k lines of engine code in a ~2.8k-line pure-Python package, MIT, on PyPI as corecoder).

Why it fits this list: CoreCoder speaks the OpenAI-compatible API natively, so DeepSeek works with three environment variables (OPENAI_API_KEY / OPENAI_BASE_URL=https://api.deepseek.com / CORECODER_MODEL=deepseek-v4-pro), and DeepSeek's reasoning_content thinking stream is handled first-class (shown dimmed, never fed back into history). The engine's provider layer, retry/restitch logic, and cost accounting are the subject of one of its eight source-reading essays, so the guide doubles as a teaching path for "how an agent talks to DeepSeek".

The PR follows the repo checklist: bilingual guides (docs/corecoder.md + docs/corecoder.zh-CN.md) in one PR, alphabetical table rows in both READMEs, current model names only (deepseek-v4-pro, no deprecated deepseek-chat), the 1M context window wired via CORECODER_MAX_CONTEXT=1000000, and an honest note that CoreCoder doesn't send reasoning_effort yet.

I'm the author of CoreCoder.

Originally created by @he-yufeng on GitHub (Oct 19, 2026). Original GitHub issue: https://github.com/deepseek-ai/awesome-deepseek-agent/issues/465 I'd like to add an integration guide for [CoreCoder](https://github.com/he-yufeng/CoreCoder), a minimal, fully readable coding-agent engine (~1.3k lines of engine code in a ~2.8k-line pure-Python package, MIT, on PyPI as `corecoder`). Why it fits this list: CoreCoder speaks the OpenAI-compatible API natively, so DeepSeek works with three environment variables (`OPENAI_API_KEY` / `OPENAI_BASE_URL=https://api.deepseek.com` / `CORECODER_MODEL=deepseek-v4-pro`), and DeepSeek's `reasoning_content` thinking stream is handled first-class (shown dimmed, never fed back into history). The engine's provider layer, retry/restitch logic, and cost accounting are the subject of one of its eight source-reading essays, so the guide doubles as a teaching path for "how an agent talks to DeepSeek". The PR follows the repo checklist: bilingual guides (`docs/corecoder.md` + `docs/corecoder.zh-CN.md`) in one PR, alphabetical table rows in both READMEs, current model names only (`deepseek-v4-pro`, no deprecated `deepseek-chat`), the 1M context window wired via `CORECODER_MAX_CONTEXT=1000000`, and an honest note that CoreCoder doesn't send `reasoning_effort` yet. I'm the author of CoreCoder.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/awesome-deepseek-agent#134
No description provided.