@deepseek-ai/dsh-tool-call-timeout-policy 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/guard/timeout-policy/README.md
-README.md: 53b298d047098cf196157ea32eaf0c2713c6daeb
-README.zh.md: 39d1914135f80fb73d4be41eb9123f6eed56bd3e
+README.md: d1499e4af072079e9be96c45582b473cd10fa5d5
+README.zh.md: f91343a42c889b04f9c772fea92c10c6ba8fe477

package/README.md

@@ -9,7 +9,7 @@ English | [中文](README.zh.md)
 
 ## Summary
 
-A tool call can hang for a long time — a slow web fetch, a search that never returns — and without a limit the model waits indefinitely, stalling the whole session. `dsh-tool-call-timeout-policy` arms a cooperative deadline for calls that declare a limit: it asks the tool to stop through `exec.signal`, then maps a settled cancellation to a clear `Error: tool call timed out after <ms>ms` result. A tool that ignores or slowly handles cancellation keeps the caller waiting until it settles; the plugin never hard-stops downstream work. The limit comes from each tool's own configuration, so the plugin itself is zero-config, and it ships enabled in the `dsh` base bundle.
+Use this package to give tool calls their configured cooperative time limits and return a clear timeout error to the model after cancellation settles. Calls that finish in time are unchanged. A tool that ignores or slowly handles cancellation can keep the caller waiting because the package cannot hard-stop downstream work. Each tool supplies its own limit; the package has no configuration and is enabled in the `dsh` base bundle.
 
 ## Table of Contents
 

package/README.zh.md

@@ -9,7 +9,7 @@ kind: "package-reference"
 
 ## 概述
 
-工具调用可能会长时间挂起——缓慢的网页抓取、永不返回的搜索——没有上限时模型会无限期等待，拖住整个会话。`dsh-tool-call-timeout-policy` 为声明了限时的调用设置协作式截止时间：它通过 `exec.signal` 请求工具停止，再把已经完成的取消映射为清晰的 `Error: tool call timed out after <ms>ms` 结果。忽略或缓慢处理取消的工具会让调用方继续等待，直到自身完成；本插件绝不会硬性停止下游工作。限时来自每个工具自身的配置，因此插件本身零配置，并随 `dsh` base 组合默认启用。
+使用本包可为工具调用执行其配置的协作式时间上限，并在取消完成后向模型返回清晰的超时错误。按时完成的调用保持不变。忽略或缓慢处理取消的工具仍可能让调用方继续等待，因为本包无法硬性停止下游工作。每个工具分别提供自己的限时；本包无需配置，并随 `dsh` base 组合默认启用。
 
 ## 目录
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@deepseek-ai/dsh-tool-call-timeout-policy",
   "description": "Tool-call timeout policy: a tools/execute wrapper that arms a per-tool deadline on exec.signal and returns TOOL_TIMEOUT when it wins",
-  "version": "0.1.3-alpha.2",
+  "version": "0.1.5-alpha.1",
   "publishConfig": {
     "access": "public"
   },
@@ -27,15 +27,15 @@
   ],
   "license": "MIT",
   "peerDependencies": {
-    "@deepseek-ai/dsh-llm": "^0.1.3-alpha.2",
-    "@deepseek-ai/dsh-timeout": "^0.1.3-alpha.2",
-    "@deepseek-ai/dsh-tools": "^0.1.3-alpha.2",
+    "@deepseek-ai/dsh-llm": "^0.1.5-alpha.1",
+    "@deepseek-ai/dsh-timeout": "^0.1.5-alpha.1",
+    "@deepseek-ai/dsh-tools": "^0.1.5-alpha.1",
     "@deepseek-ai/cordis": "^4.0.2"
   },
   "devDependencies": {
-    "@deepseek-ai/dsh-llm": "^0.1.3-alpha.2",
-    "@deepseek-ai/dsh-timeout": "^0.1.3-alpha.2",
-    "@deepseek-ai/cordis": "^4.0.2",
-    "@deepseek-ai/dsh-tools": "^0.1.3-alpha.2"
+    "@deepseek-ai/dsh-llm": "^0.1.5-alpha.1",
+    "@deepseek-ai/dsh-timeout": "^0.1.5-alpha.1",
+    "@deepseek-ai/dsh-tools": "^0.1.5-alpha.1",
+    "@deepseek-ai/cordis": "^4.0.2"
   }
 }