[PR #457] Add AgenC Core guide #451

Open
opened 2026-10-26 09:49:41 +00:00 by bsakaguchi · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-agent/pull/457
Author: @signerless
Created: 10/16/2026
Status: 🔄 Open

Base: main ← Head: add-agenc-core


📝 Commits (1)

📊 Changes

4 files changed (+288 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
📝 README.zh-CN.md (+1 -0)
➕ docs/agenc.md (+143 -0)
➕ docs/agenc.zh-CN.md (+143 -0)

📄 Description

Adds AgenC Core, an open-source (MIT) terminal coding agent with a built-in deepseek provider. Guides in English and Chinese (docs/agenc.md, docs/agenc.zh-CN.md) and a row in both READMEs, first in alphabetical order (before AstrBot).

See CONTRIBUTING.md for full guidelines.

Checklist

Agent Configuration

  • Model names are current (v4-pro / v4-flash, not v3 names)
  • 1M context is configured or mentioned
  • Max thinking effort level is supported
  • Pricing is current (input, output, cache hit), verified against DeepSeek API Docs
  • Config fields are verified to work in the agent
  • No workarounds for already-fixed upstream bugs

Documentation

  • Both English and Chinese versions included in a single PR
  • README table entry inserted in alphabetical order
  • One tool per PR; unrelated tools not combined
  • Images placed in docs/assets/ with descriptive filenames and reasonable sizes (no images)

Notes for review

  • Model IDs. The guide uses deepseek-flash (DeepSeek-V4.1-Flash) and deepseek-v4-pro, the two IDs on DeepSeek's Models & Pricing and List Models pages as of 2026-10-16. The pricing page footnote says to use deepseek-flash and that deepseek-v4-flash still works; the guide mentions both. Open PR #418 moves the rest of the list to deepseek-flash too.
  • 1M context. AgenC's catalog sets 1,048,576 tokens for both models, so no config is needed. The guide shows the override keys for proxies.
  • Max effort. AgenC sends thinking: {"type": "enabled"} and reasoning_effort: "max" unchanged, and replays reasoning_content for tool calls as the Thinking Mode docs require.
  • Pricing. The guide quotes no prices and links to DeepSeek's pricing page, so it cannot go stale.
  • Config. agenc config set model_provider deepseek, model deepseek-flash and reasoning_effort max were checked with agenc 0.19.6 and pass agenc config validate. The same model and effort ran the public Terminal-Bench 2.1 job linked in the guide.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/awesome-deepseek-agent/pull/457 **Author:** [@signerless](https://github.com/signerless) **Created:** 10/16/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `add-agenc-core` --- ### 📝 Commits (1) - [`4b1ab02`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/4b1ab0204ed6b7a447277fa3d189f5bef151ef01) Add AgenC Core guide ### 📊 Changes **4 files changed** (+288 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) 📝 `README.zh-CN.md` (+1 -0) ➕ `docs/agenc.md` (+143 -0) ➕ `docs/agenc.zh-CN.md` (+143 -0) </details> ### 📄 Description Adds [AgenC Core](https://github.com/tetsuo-ai/agenc-core), an open-source (MIT) terminal coding agent with a built-in `deepseek` provider. Guides in English and Chinese (`docs/agenc.md`, `docs/agenc.zh-CN.md`) and a row in both READMEs, first in alphabetical order (before AstrBot). See [CONTRIBUTING.md](./CONTRIBUTING.md) for full guidelines. ## Checklist ### Agent Configuration - [x] Model names are current (v4-pro / v4-flash, not v3 names) - [x] 1M context is configured or mentioned - [x] Max thinking effort level is supported - [x] Pricing is current (input, output, cache hit), verified against [DeepSeek API Docs](https://api-docs.deepseek.com/quick_start/pricing) - [x] Config fields are verified to work in the agent - [x] No workarounds for already-fixed upstream bugs ### Documentation - [x] Both English and Chinese versions included in a single PR - [x] README table entry inserted in alphabetical order - [x] One tool per PR; unrelated tools not combined - [x] Images placed in `docs/assets/` with descriptive filenames and reasonable sizes (no images) ## Notes for review - **Model IDs.** The guide uses `deepseek-flash` (DeepSeek-V4.1-Flash) and `deepseek-v4-pro`, the two IDs on DeepSeek's [Models & Pricing](https://api-docs.deepseek.com/quick_start/pricing) and [List Models](https://api-docs.deepseek.com/api/list-models) pages as of 2026-10-16. The pricing page footnote says to use `deepseek-flash` and that `deepseek-v4-flash` still works; the guide mentions both. Open PR #418 moves the rest of the list to `deepseek-flash` too. - **1M context.** AgenC's catalog sets 1,048,576 tokens for both models, so no config is needed. The guide shows the override keys for proxies. - **Max effort.** AgenC sends `thinking: {"type": "enabled"}` and `reasoning_effort: "max"` unchanged, and replays `reasoning_content` for tool calls as the Thinking Mode docs require. - **Pricing.** The guide quotes no prices and links to DeepSeek's pricing page, so it cannot go stale. - **Config.** `agenc config set model_provider deepseek`, `model deepseek-flash` and `reasoning_effort max` were checked with `agenc 0.19.6` and pass `agenc config validate`. The same model and effort ran the public Terminal-Bench 2.1 job linked in the guide. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
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#451
No description provided.