@deepseek-ai/dsh-llm-mock-server 0.1.3-alpha.2 → 0.1.5-alpha.1

package/README.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/test-support/llm-mock-server/README.md
-README.md: 39e41f47469e06f1638a8c90ca5f9d0ce46a4d67
-README.zh.md: 983522beb50236acc3e94cee84bb95b1e591e75f
+README.md: be311d9c83bcd35ad5bd0c003f6e95861b6d6976
+README.zh.md: 4c8dd1fffcb38be36d6e67420fa6be2370b89484

package/README.md

@@ -9,7 +9,7 @@ English | [中文](README.zh.md)
 
 ## Summary
 
-`dsh-llm-mock-server` stands in for a real model provider during tests as a scriptable OpenAI-compatible HTTP/SSE server: you script a sequence of wire behaviors — stream resets, stalls, malformed chunks, rate limits, server errors, successful completions, tool calls — and each accepted `/chat/completions` request consumes the next one. It serves the shipping DeepSeek adapter and the agent loop over real HTTP, so recovery policy such as retries, backoff, and timeouts is exercised against a genuine wire boundary without a provider key. A CLI (`pnpm run mock:llm`) runs the server standalone; the library entry `startMockLlmServer` embeds it in tests and returns captured requests. A `random` behavior with seeded weights mixes failures for open-ended stress runs.
+This package gives tests and demos a scriptable OpenAI-compatible HTTP/SSE endpoint, so they can exercise model-provider failures and successes without a provider key. Each accepted `/chat/completions` request consumes the next scripted behavior, including resets, stalls, malformed chunks, rate limits, server errors, completions, and tool calls. Test authors can run it with `pnpm run mock:llm` or call `startMockLlmServer`, which returns captured requests for assertions. Seeded `random` behavior supports reproducible mixed-failure stress runs.
 
 ## Table of Contents
 

package/README.zh.md

@@ -9,7 +9,7 @@ kind: "package-library"
 
 ## 概述
 
-`dsh-llm-mock-server` 在测试期间以可编脚本的 OpenAI 兼容 HTTP／SSE（Server-Sent Events）服务器代替真实模型提供方：你脚本化一串协议行为——流重置、停滞、畸形分片、限流、服务器错误、成功补全、工具调用——每个已接受的 `/chat/completions` 请求依次消费下一个。它通过真实 HTTP 服务发布的 DeepSeek 适配器与 agent loop（智能体循环），因此重试、退避与超时等恢复策略会在真实协议边界上得到检验，且无需提供方密钥。CLI（`pnpm run mock:llm`）可独立运行服务器；库入口 `startMockLlmServer` 将其嵌入测试并返回捕获的请求。`random` 行为配合带种子的权重可混合故障，用于开放式压力运行。
+本包为测试与演示提供可编脚本的 OpenAI 兼容 HTTP／SSE 端点，使其无需提供方密钥即可检验模型提供方的失败与成功。每个已接受的 `/chat/completions` 请求依次消费下一个脚本行为，包括重置、停滞、畸形分片、限流、服务器错误、补全与工具调用。测试作者可以通过 `pnpm run mock:llm` 运行服务器，也可以调用 `startMockLlmServer`，后者会返回捕获的请求供断言使用。带种子的 `random` 行为支持可复现的混合故障压力运行。
 
 ## 目录
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@deepseek-ai/dsh-llm-mock-server",
   "description": "Scriptable OpenAI-compatible HTTP/SSE fault server for LLM recovery tests",
-  "version": "0.1.3-alpha.2",
+  "version": "0.1.5-alpha.1",
   "publishConfig": {
     "access": "public"
   },