billion-context 0.1.105 → 0.1.106-pr.711.554
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +10 -0
- package/README.zh-CN.md +10 -0
- package/dist/index.js +366 -197
- package/dist/index.js.map +1 -1
- package/dist/mcp.js.map +1 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -24,6 +24,12 @@ Any agent that can set a base URL — <em>zero per-agent adapter code</em>.
|
|
|
24
24
|
|
|
25
25
|
`billion-context` sits between **any** agent and its model API, rewriting Anthropic/OpenAI streams with [acp-kernel](https://github.com/ranxianglei/acp-kernel) compression. The model decides **when** and **what** to compress into high-fidelity summaries — not a hard truncation limit.
|
|
26
26
|
|
|
27
|
+
## Community
|
|
28
|
+
|
|
29
|
+
Discussion, help, and updates on QQ — one group covers all three projects (`billion-context`, `billion-context-pi`, `opencode-acp`):
|
|
30
|
+
|
|
31
|
+
**QQ Group: 1056132097**
|
|
32
|
+
|
|
27
33
|
## Why
|
|
28
34
|
|
|
29
35
|
Long coding sessions blow up context. Each provider charges per token, and once you pass the context window the session degrades or dies. `billion-context` compresses consumed conversation into layered summaries so you can run a single session for days — billions of tokens through one context window.
|
|
@@ -466,6 +472,10 @@ Early. Protocol handling and compression work against mock tests (500+ passing).
|
|
|
466
472
|
|
|
467
473
|
Client-side plugins for pi / omp / opencode ship inside `billion-context` (`dist/agent/*.js`) for the cooperative-proxy path. See the **"Which do I need?"** section above for how `billion-context`, the standalone `billion-context-pi`, and `opencode-acp` relate.
|
|
468
474
|
|
|
475
|
+
## Community
|
|
476
|
+
|
|
477
|
+
QQ group — one shared group for all three projects ([`billion-context`](https://github.com/ranxianglei/billion-context), [`billion-context-pi`](https://github.com/ranxianglei/billion-context-pi), [`opencode-acp`](https://github.com/ranxianglei/opencode-acp)): **1056132097**
|
|
478
|
+
|
|
469
479
|
## License
|
|
470
480
|
|
|
471
481
|
MIT
|
package/README.zh-CN.md
CHANGED
|
@@ -24,6 +24,12 @@ AI 编程助手的<strong>通用上下文压缩代理</strong>
|
|
|
24
24
|
|
|
25
25
|
`billion-context` 架在**任意**编程助手与其模型 API 之间,用 [acp-kernel](https://github.com/ranxianglei/acp-kernel) 压缩重写 Anthropic/OpenAI 流。何时压缩、压缩什么 —— <strong>由模型决定</strong>,而非硬截断。
|
|
26
26
|
|
|
27
|
+
## 社区
|
|
28
|
+
|
|
29
|
+
交流、求助与更新都在 QQ——同一个群覆盖三个项目(`billion-context`、`billion-context-pi`、`opencode-acp`):
|
|
30
|
+
|
|
31
|
+
**QQ 群:1056132097**
|
|
32
|
+
|
|
27
33
|
## 为什么
|
|
28
34
|
|
|
29
35
|
长编程会话会把上下文撑爆。各家 provider 按 token 计费,一旦超过上下文窗口,会话质量下降甚至崩掉。`billion-context` 把已消耗的对话压缩成分层摘要,让你**一个会话连跑数天** —— 海量 token 穿过同一个上下文窗口。
|
|
@@ -287,6 +293,10 @@ Windows 下会自动发现常见 Clash/Mihomo 静态系统代理;Web UI 会显
|
|
|
287
293
|
|
|
288
294
|
针对 pi / omp / opencode 的客户端插件随 `billion-context` 一起发布(`dist/agent/*.js`),用于协作代理路径。三者(`billion-context`、独立的 `billion-context-pi`、`opencode-acp`)如何取舍,见上文「该选哪个?」一节。
|
|
289
295
|
|
|
296
|
+
## 社区
|
|
297
|
+
|
|
298
|
+
QQ 群 —— 三个项目共用一个群([`billion-context`](https://github.com/ranxianglei/billion-context)、[`billion-context-pi`](https://github.com/ranxianglei/billion-context-pi)、[`opencode-acp`](https://github.com/ranxianglei/opencode-acp)):**1056132097**
|
|
299
|
+
|
|
290
300
|
## 许可证
|
|
291
301
|
|
|
292
302
|
MIT
|