deepagent-ltd · deepagent-ai · Aug 8, 2026 · Aug 6, 2026 · Aug 8, 2026
diff --git a/.github/workflows/deepagent-code.yml b/.github/workflows/deepagent-code.yml
@@ -26,7 +26,7 @@ jobs:
       - uses: ./.github/actions/setup-bun
 
       - name: Run deepagent-code
-        uses: deepagent-ltd/deepagent-code/github@github-v1.4.4
+        uses: deepagent-ltd/deepagent-code/github@github-v1.4.5
         env:
           DEEPAGENT_CODE_API_KEY: ${{ secrets.DEEPAGENT_CODE_API_KEY }}
           DEEPAGENT_CODE_PERMISSION: '{"bash": "deny"}'

diff --git a/.github/workflows/docs-update.yml b/.github/workflows/docs-update.yml
@@ -43,7 +43,7 @@ jobs:
 
       - name: Run deepagent-code
         if: steps.commits.outputs.has_commits == 'true'
-        uses: deepagent-ltd/deepagent-code/github@github-v1.4.4
+        uses: deepagent-ltd/deepagent-code/github@github-v1.4.5
         env:
           DEEPAGENT_CODE_API_KEY: ${{ secrets.DEEPAGENT_CODE_API_KEY }}
         with:

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,14 @@
 # Changelog
 
+## Desktop 1.4.5 / DeepAgent Core V4.0.6 - Bounded planning and runtime hardening
+
+- Make compaction durable through explicit run/attempt lifecycle records and Prompt Epoch authority, with bounded text-only summary retries and typed terminal failures.
+- Reject unknown context limits before unsafe dispatch and compare raw-token phases without treating missing provider metadata as a known capacity.
+- Add readiness-aware context federation and per-turn provider request/argument receipts across adapter, AI SDK, processor, and available raw-frame boundaries.
+- Enforce atomic plan admission, versioned compare-and-commit, last-known-good preservation, bounded malformed-plan attempts, durable human edits, and cursor-safe UI reconciliation.
+- Verify the compiled and installed package with the typed plan HTTP matrix and a DeepSeek malformed-recovery plus happy-path plan run.
+- Align Desktop 1.4.5, DeepAgent Core V4.0.6, GitHub Action `github-v1.4.5`, package metadata, updater tags, workflow references, and public release documentation.
+
 ## Desktop 1.4.4 / DeepAgent Core V4.0.5 - Durable context and collaboration
 
 - Persist AgentExecution claims, leases, generations, resource locks, token debits, terminal metadata, and handoffs for event-driven multi-Agent work.

diff --git a/README.md b/README.md
@@ -14,8 +14,6 @@
   <a href="https://github.com/deepagent-ltd/deepagent-code-enterprise">Enterprise</a>
 </p>
 
-<p align="center"><sub>Desktop 1.4.4 · DeepAgent Core V4.0.5</sub></p>
-
 ---
 
 DeepAgent Code is an AI coding workspace for work that lasts longer than one prompt. It combines a production coding-agent runtime with durable sessions, connected project memory, live planning, code intelligence, multi-agent collaboration, and human oversight.
@@ -93,19 +91,6 @@ For high-risk decisions, convene an **Expert Panel**. Correctness, security, per
 
 Project IM brings people and agents into the same thread. Mention an agent to start a scoped run with project context, stream its progress, inspect its artifacts, and keep the answer attached to the conversation that requested it.
 
-## DeepAgent Core V4.0.5
-
-Desktop 1.4.4 ships the durable context and collaboration release. It connects Session V2, federated context, delegated execution, Git/PR review, and human supervision without moving durable authority back into process-local agent loops:
-
-- **Durable context and execution:** prompt admission, Context Epoch selection, AgentExecution claims, leases, generations, resource locks, token debits, terminal metadata, and handoffs are persisted before scheduling provider work.
-- **Complete multi-agent Git/PR collaboration:** write workers use canonical isolated worktrees, commit only scoped changes, and enqueue SHA-bound PRs. A normal Reviewer handles each item, merges are serialized with `--no-ff`, and a Senior Reviewer closes the batch.
-- **Generation-fenced supervision:** resume, cancellation, timeout, takeover, review feedback, and cleanup reuse durable Session identity while preventing stale workers from settling newer work.
-- **One private filesystem root:** Core, CLI, Desktop, WSL, databases, logs, caches, and temporary files stay under `~/.deepagent/code/`; production writes cannot be redirected by ordinary environment variables.
-- **Cache-safe real-model verification:** stable prompt prefixes are separated from volatile tails, finalizer and compaction baselines are isolated, and the DeepSeek matrix covers sandboxed tools, cache retention, parallel workers, PR closure, recovery, strength inheritance, and EVAL scoring.
-- **Release-grade verification:** deterministic state-machine tests, cross-process ownership tests, real-model suites, packaged sidecar checks, Desktop UI coverage, and source-map smoke tests exercise the production entry points.
-
-V4.0.5 retains the durable documents, event delivery, governed learning, secure credential boundaries, and fail-closed provider/tool contracts established in the V4.0 line.
-
 ## Installation
 
 > **Note:** The `deepagent-code` npm package is not yet publicly published.

diff --git a/README.zh.md b/README.zh.md
@@ -14,8 +14,6 @@
   <a href="https://github.com/deepagent-ltd/deepagent-code-enterprise">Enterprise 版本</a>
 </p>
 
-<p align="center"><sub>桌面版 1.4.4 · DeepAgent Core V4.0.5</sub></p>
-
 ---
 
 DeepAgent Code 是一套面向长期工作的 AI 编程工作区。它把生产级编程智能体运行时与持久会话、相互连通的项目记忆、实时计划、代码智能、多智能体协作和人类监督组合在一起。
@@ -93,19 +91,6 @@ DeepAgent 可以把独立工作拆分给数量有界、相互隔离的 Worker。
 
 项目 IM 把团队成员和智能体放进同一条讨论。@ 某个智能体即可启动有明确作用域的运行，使用项目上下文、流式展示进度、关联执行工件，并把答案留在发起任务的对话里。
 
-## DeepAgent Core V4.0.5
-
-桌面版 1.4.4 搭载持久上下文与协作版本。它把 Session V2、联邦上下文、委派执行、Git/PR 审阅和人类监督连接起来，同时不把持久权威退回到进程内 Agent loop：
-
-- **持久上下文与执行：** prompt 准入、Context Epoch 选择、AgentExecution claim、lease、generation、资源锁、token debit、终态 metadata 和 handoff 都在调度 provider work 前持久化。
-- **完整多 Agent Git/PR 协作：** 写 Worker 使用 canonical 隔离 worktree，只提交作用域内改动，并进入绑定精确 SHA 的 PR queue。普通 Reviewer 逐项审阅，协调器串行执行 `--no-ff` merge，再由 Senior Reviewer 完成批次复审。
-- **受 generation fence 保护的监督：** 恢复、取消、超时、接管、审阅反馈和清理复用持久 Session 身份，同时阻止过期 Worker 结算较新的工作。
-- **唯一私有文件数据根：** Core、CLI、Desktop、WSL、数据库、日志、缓存和临时文件统一位于 `~/.deepagent/code/`；普通环境变量不能重定向生产写入。
-- **缓存安全的真实模型验证：** 稳定 prompt prefix 与易变 tail 分离，finalizer 和 compaction 使用隔离基线；DeepSeek 矩阵覆盖 sandbox 工具、缓存保持、并行 Worker、PR 闭环、恢复、强度继承与 EVAL 评分。
-- **发布级验证：** 确定性状态机测试、跨进程 ownership 测试、真实模型 suite、packaged sidecar 检查、Desktop UI 覆盖和 source-map smoke 共同验证生产入口。
-
-V4.0.5 保留 V4.0 系列建立的持久文档、可靠事件投递、知识治理、安全凭据边界以及失败关闭的 provider/tool 合同。
-
 ## 安装
 
 > **说明：** `deepagent-code` npm 包尚未公开发布。

diff --git a/bun.lock b/bun.lock
diff --git a/design/README.md b/design/README.md
@@ -1,6 +1,6 @@
 # DeepAgent Code Architecture & Design
 
-> **Public design overview for DeepAgent Core V4.0.5 / Desktop 1.4.4.** Internal implementation details and roadmap documents live in the private `docs/` tree and are intentionally not version-controlled.
+> **Public architecture overview.** Internal implementation details and roadmap documents live in the private `docs/` tree and are intentionally not version-controlled.
 
 DeepAgent Code is a document-centered, event-driven AI coding system. It combines a coding-agent runtime with a durable control plane that owns context, planning, learning, collaboration, safety, and human oversight.
 
@@ -26,7 +26,7 @@ Sessions, inputs, plans, documents, goals, events, approvals, and learning decis
 
 A user instruction is durably admitted before execution is scheduled. A successful API response therefore means the instruction is recorded, not merely present in a process-local queue.
 
-DeepAgent is built **on top of** the opencode agent/runtime/session/tool/MCP stack. V4.0.5 strengthens the durable control plane without replacing the current turn engine, tool system, or provider layer.
+DeepAgent is built **on top of** the opencode agent/runtime/session/tool/MCP stack. Its durable control plane does not replace the current turn engine, tool system, or provider layer.
 
 ### Durable delegated execution
 

diff --git a/design/real-llm-testing.md b/design/real-llm-testing.md
@@ -6,7 +6,7 @@
 
 ## 0. 权威状态与硬合同
 
-截至 2026-07-31，聚合 runner 注册 55 条命令，其中 48 条调用真实模型；headless 矩阵为 48 条命令、43 条真实模型 suite，进一步跳过 EVAL 和安装后为 46 条命令、42 条真实模型 suite。数字由 `script/run-live-llm-all.ts` 动态注册表决定，文档数字发生漂移时以注册表和 `validateSuiteManifest()` 为准。
+截至 2026-08-06，聚合 runner 注册 58 条命令，其中 51 条调用真实模型；headless 矩阵为 51 条命令、46 条真实模型 suite，进一步跳过 EVAL 和安装后为 49 条命令、45 条真实模型 suite。数字由 `script/run-live-llm-all.ts` 动态注册表决定，文档数字发生漂移时以注册表和 `validateSuiteManifest()` 为准。
 
 `qualifiedLiveRuns` 当前为空。注册、单次通过和 EXT 可达都不代表 pre-push 资格；LIVE suite 只有完成本节资格合同后才能进入该集合。
 
@@ -127,6 +127,14 @@ ${EDITOR:-vi} script/live-llm.config.local.json
 | `evalRuns`            | EVAL 重复次数，范围为 1–20。次数越高，费用和耗时越高。               |
 | `installDependencies` | 聚合测试是否先运行依赖安装步骤。命令行 `--skip-install` 可以覆盖它。 |
 
+旧版配置如果仍包含 `"apiKey": "..."`，必须迁移，不能同时保留两个字段：
+
+1. 将原始 key 移入第 2 节所述的仓库外单行文件，并执行 `chmod 600`；
+2. 删除 JSON 中的 `apiKey`；
+3. 在同一位置写入 `"apiKeyFile": "~/.deepagent/code/tmp/live-llm-deepseek.key"`。
+
+聚合 runner 会在任何 Provider 请求发生前拒绝旧字段，且错误信息不会回显旧 key。package 和 Desktop 的单 suite 入口同样只接受 `DEEPAGENT_CODE_LIVE_LLM_API_KEY_FILE`。
+
 ## 4. 启动聚合测试
 
 所有聚合命令都从仓库根目录运行。
@@ -200,6 +208,7 @@ export DEEPAGENT_CODE_LIVE_LLM_TIMEOUT_MS="180000"
 (cd packages/deepagent-code && bun run test:llm-ext:subagent-intensity)
 (cd packages/deepagent-code && bun run test:llm-ext:subagent-resume)
 (cd packages/deepagent-code && bun run test:llm-ext:subagent-takeover)
+(cd packages/deepagent-code && bun run test:llm-ext:prompt-intent-fencing)
 (cd packages/desktop && bun run test:llm-release:ui)
 ```
 
@@ -243,7 +252,13 @@ Git 合同分入口定义：交互式写型 `task` 使用独立 worktree，并
 
 审批后的完整闭环由直接生产状态机测试验证：Reviewer 首次返回 `request_changes` 后，使用原 `task_id` 恢复同一个 V4 child Session 和作者 worktree，重新提交到同一 PR；再次 `pr_finalize` 后绑定新 SHA 审阅、`--no-ff` merge、Senior Reviewer 和 worktree cleanup 全部完成。默认 V4 分区规则当前没有产生独立同 wave 节点，因此 V4 wave pool 的并行性使用确定性双 runner 同步屏障证明；真实模型并行性由 `multi-agent-parallel-worktrees` 和 `expert-panel` 覆盖，不能把两类证据混写。
 
-### 5.5 Expert Panel 与缓存边界
+### 5.5 智能直达、Prompt Intent 与撤回重写
+
+`prompt-intent-fencing` 在同一个真实 DeepSeek Session 中覆盖两条生产链。第一条先通过 intelligence prepare 得到 `route=general`，再以同一个 durable intent 提交原始输入；Oracle 要求 `session_intent` 为 `admitted/original/turn`、Session 中只有一条原始用户消息和一次真实 provider 执行。在 provider 仍为 busy 的窗口内，ACK 丢失后的 exact retry 不得增加用户消息，晚到的 rewritten 草稿必须以 `SessionPromptIntent.Conflict` 拒绝，不得排队或启动第二次执行。第二条对该消息执行生产 `SessionRevert`，要求 mutation epoch 精确递增一次，旧 intent retry 以 `SessionMutationEpoch.Stale` 拒绝，再以新 rewrite intent 完成一次真实 provider 执行；清理后的活动 transcript 只能保留一条 rewrite 用户消息，且不得残留旧回复、工具调用、权限请求或工作区改动。
+
+该 EXT suite 只证明真实 refinement/provider 与 durable admission/revert 接线一致，不替代竞态正确性证明。并发 claim、事务原子性、ACK crash window、跨窗口相同 intent、revert 与 materialize 的交错顺序仍由 SQLite/Effect/前端 barrier 的确定性测试承担；不得用模型最终文本或单次 live 通过声称 exactly-once 状态机已经完备。
+
+### 5.6 Expert Panel 与缓存边界
 
 `expert-panel` 必须并行创建 correctness、security、architecture 三个 Reviewer Session。每个 Reviewer 只能成功读取自己的目标文件，必须经过独立的 structured finalizer，并由确定性 arbiter 重算最终 verdict 和 dissent。Oracle 要求每个 lens 至少有一条命中预埋代码事实且置信度为 `0.95` 的 finding；额外 finding 只要仍受目标文件、类别、证据隔离和置信度范围约束，就不应被误判为产品故障。模型对未授权路径的额外 read 尝试可以存在，但必须由权限层拒绝；任何额外成功读取或其他工具调用都应失败。
 

diff --git a/github/README.md b/github/README.md
@@ -2,7 +2,7 @@
 
 A GitHub Action that integrates [deepagent-code](https://deepagent-code.ai) directly into your GitHub workflow.
 
-Current Action release: `github-v1.4.4`. The release workflow reads the target version from `github/VERSION` and publishes that exact tag.
+The release workflow reads the target version from `github/VERSION` and publishes that exact tag.
 
 Mention `/deepagent-code` in your comment, and deepagent-code will execute tasks within your GitHub Actions runner.
 
@@ -90,7 +90,7 @@ This will walk you through installing the GitHub app, creating the workflow, and
               persist-credentials: false
 
           - name: Run deepagent-code
-            uses: deepagent-ltd/deepagent-code/github@github-v1.4.4
+            uses: deepagent-ltd/deepagent-code/github@github-v1.4.5
             env:
              ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
              GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

diff --git a/github/VERSION b/github/VERSION
@@ -1 +1 @@
-1.4.4
+1.4.5
diff --git a/packages/app/README.md b/packages/app/README.md
@@ -2,8 +2,6 @@
 
 SolidJS front-end shell for the DeepAgent Code desktop app (Electron/Tauri).
 
-Current release: Desktop 1.4.4, powered by DeepAgent Core V4.0.5.
-
 ## Stack
 
 - **UI:** SolidJS + Vite (Bun)

diff --git a/packages/app/package.json b/packages/app/package.json
@@ -1,6 +1,6 @@
 {
   "name": "@deepagent-code/app",
-  "version": "1.4.4",
+  "version": "1.4.5",
   "description": "",
   "type": "module",
   "exports": {