Merge pull request 'Add bounded evidence to work checkpoint candidates' (#31 ) from feature/fast-slow-u2-u3 into main

Merge the fast-slow U2/U3 branch history after syncing with main
agent-exec transport 68889ca8d4b0eec4cfe89d0cc80b6516/5ca3b7798211
2026-10-19 09:20:24 +00:00 · 2026-10-19 02:59:18 +00:00 · 2026-10-19 02:59:03 +00:00 · 2026-10-17 08:31:36 +00:00 · 2026-10-17 08:25:12 +00:00 · 2026-10-17 07:15:37 +00:00
25 changed files with 1125 additions and 61 deletions
@@ -0,0 +1,150 @@
+name: governance
+on:
+  pull_request:
+  push:
+    branches: [main]
+
+jobs:
+  governance:
+    runs-on: ubuntu-latest
+    env:
+      GOPROXY: https://goproxy.cn,direct
+    steps:
+      - name: Checkout
+        env:
+          TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          SUBMODULE_TOKEN: ${{ secrets.SUBMODULE_TOKEN }}
+        run: |
+          set -euo pipefail
+          # 只物化两个治理 gitlink（Guidelines 与 deps/development-manager），非递归；
+          # 其它 gitlink 不物化、不取回。凭据只以 URL 限定的 extraheader 存在于本进程树内。
+          python3 - <<'PY_CHECKOUT'
+          import os
+          from pathlib import Path
+          import re
+          import subprocess
+          import sys
+          from urllib.parse import urlsplit
+
+          # 服务端基址取运行器注入的 GITHUB_SERVER_URL（平台控制、候选 PR 不可影响）。
+          SERVER = os.environ.get("GITHUB_SERVER_URL", "").rstrip("/")
+          parsed_server = urlsplit(SERVER)
+          if parsed_server.scheme not in {"http", "https"} or not parsed_server.netloc:
+              sys.exit("Checkout server must be the platform-injected http(s) URL")
+          SSH = "ssh://git@git-ssh.u1680832.nyat.app:31411/weichen/"
+          # 审查过的白名单，独立于候选 .gitmodules：仅列出两个治理 gitlink。
+          REPOSITORIES = {
+              "deps/development-guidelines": "Development-Guidelines",
+              "deps/development-manager": "Development-Manager",
+          }
+          TARGET = "weichen/Agent-System-Core"
+          root_token = os.environ.pop("TOKEN", "")
+          component_token = os.environ.pop("SUBMODULE_TOKEN", "")
+          if not root_token or not component_token:
+              sys.exit("Checkout requires TOKEN and SUBMODULE_TOKEN")
+          if os.environ.get("GITHUB_REPOSITORY") != TARGET:
+              sys.exit("Checkout repository is outside the reviewed allowlist")
+          sha = os.environ.get("GITHUB_SHA", "")
+          if not re.fullmatch(r"[0-9a-fA-F]{40}", sha):
+              sys.exit("Checkout requires a fixed commit SHA")
+
+          # 丢弃继承的 Git 追踪/配置/SSH 钩子；凭据只存在于单个 Git 进程树的精确 URL 头中。
+          clean_env = {k: v for k, v in os.environ.items() if not k.startswith("GIT_")}
+          clean_env.update(GIT_CONFIG_NOSYSTEM="1", GIT_CONFIG_SYSTEM="/dev/null",
+                           GIT_CONFIG_GLOBAL="/dev/null", GIT_TERMINAL_PROMPT="0")
+          config = {
+              "core.hooksPath": "/dev/null", "core.fsmonitor": "false",
+              "init.templateDir": "", "credential.helper": "",
+              "fetch.recurseSubmodules": "false", "submodule.recurse": "false",
+              "http.followRedirects": "false", "protocol.allow": "never",
+              "protocol." + urlsplit(SERVER).scheme + ".allow": "always",
+          }
+
+
+          def git(repo, *args, extra=None):
+              values = config | (extra or {})
+              env = clean_env | {"GIT_CONFIG_COUNT": str(len(values))}
+              for index, (key, value) in enumerate(values.items()):
+                  env[f"GIT_CONFIG_KEY_{index}"] = key
+                  env[f"GIT_CONFIG_VALUE_{index}"] = value
+              return subprocess.run(["git", "-C", str(repo), *args], env=env,
+                                    check=True, stdout=subprocess.PIPE).stdout
+
+
+          def materialize_governance(root):
+              # -r 列出任意深度的 gitlink，但不进入 submodule 内部；其它 gitlink 只记录、从不取回。
+              links = {}
+              for entry in git(root, "ls-tree", "-rz", "HEAD").split(b"\0"):
+                  if not entry:
+                      continue
+                  metadata, path = entry.split(b"\t", 1)
+                  mode, kind, commit = metadata.decode().split()
+                  if mode == "160000":
+                      links[path.decode()] = commit
+              for name in REPOSITORIES:
+                  if name not in links:
+                      raise ValueError("Governance gitlink is missing from the fixed commit")
+              raw = git(root, "config", "--no-includes", "--blob", "HEAD:.gitmodules", "--null", "--list")
+              entries = {}
+              for item in raw.split(b"\0"):
+                  if not item:
+                      continue
+                  key, value = item.decode().split("\n", 1)
+                  match = re.fullmatch(r"submodule\.(.+)\.(path|url)", key)
+                  if not match:
+                      continue
+                  name, field = match.groups()
+                  if name not in REPOSITORIES:
+                      continue
+                  if field in entries.setdefault(name, {}):
+                      raise ValueError("Duplicate candidate .gitmodules setting")
+                  entries[name][field] = value
+              # 先整体校验两个治理条目，再进行任何带凭据的取回。
+              for name, repository in REPOSITORIES.items():
+                  fields = entries.get(name, {})
+                  if fields.get("path") != name:
+                      raise ValueError("Governance submodule path is outside the reviewed allowlist")
+                  trusted_url = f"{SERVER}/weichen/{repository}.git"
+                  if fields.get("url") not in {SSH + repository + ".git", trusted_url}:
+                      raise ValueError("Governance submodule URL is outside the reviewed allowlist")
+              for name, repository in REPOSITORIES.items():
+                  url = f"{SERVER}/weichen/{repository}.git"
+                  # 仅持久化审查过的公开 URL，使 Git 在进程级配置消失后仍识别该 submodule。
+                  git(root, "config", f"submodule.{name}.url", url)
+                  git(root, "submodule", "update", "--init", "--checkout", "--", name,
+                      extra={f"submodule.{name}.url": url,
+                             f"submodule.{name}.update": "checkout",
+                             f"http.{url}.extraheader": "AUTHORIZATION: token " + component_token})
+                  child = root / name
+                  if git(child, "rev-parse", "HEAD").decode().strip() != links[name]:
+                      raise ValueError("Materialized governance gitlink does not match the fixed SHA")
+
+
+          try:
+              root = Path.cwd()
+              # 必须是干净的临时工作区；不复用既有 Git 状态。
+              if (root / ".git").exists():
+                  raise ValueError("Checkout requires a fresh workspace")
+              git(root, "init", "-q")
+              url = f"{SERVER}/{TARGET}.git"
+              git(root, "remote", "add", "origin", url)
+              git(root, "fetch", "--depth=1", "origin", sha,
+                  extra={f"http.{url}.extraheader": "AUTHORIZATION: token " + root_token})
+              git(root, "checkout", "-q", "--detach", "FETCH_HEAD")
+              materialize_governance(root)
+          except (ValueError, OSError, subprocess.CalledProcessError) as error:
+              sys.exit(f"Checkout refused: {error}")
+          print("Checkout complete; credentials were not persisted in Git configuration")
+          PY_CHECKOUT
+          unset TOKEN SUBMODULE_TOKEN
+      - name: Install Go 1.27.1
+        run: |
+          set -euo pipefail
+          curl -fsSL -o "$RUNNER_TEMP/go.tgz" https://mirrors.aliyun.com/golang/go1.27.1.linux-amd64.tar.gz
+          tar -C /usr/local -xzf "$RUNNER_TEMP/go.tgz"
+          ln -sfn /usr/local/go/bin/go /usr/local/bin/go
+          ln -sfn /usr/local/go/bin/gofmt /usr/local/bin/gofmt
+      - name: Validate agent execution configuration
+        run: scripts/agent-exec validate --repo .
+      - name: Governance check
+        run: scripts/dm governance check --repo .
@@ -0,0 +1,6 @@
+[submodule "deps/development-guidelines"]
+	path = deps/development-guidelines
+	url = ssh://git@git-ssh.u1680832.nyat.app:31411/weichen/Development-Guidelines.git
+[submodule "deps/development-manager"]
+	path = deps/development-manager
+	url = ssh://git@git-ssh.u1680832.nyat.app:31411/weichen/Development-Manager.git
@@ -0,0 +1,53 @@
+# Agent-System-Core 采纳 development-governance-v2 治理套件 任务卡
+
+任务主记录：[Agent-System-Core PR #35](https://git-home.u1680832.nyat.app:30319/weichen/Agent-System-Core/pulls/35)（本文件随 PR 描述维护）。
+状态：已完成（闭合随本 PR 合入；合入以平台记录为准）；owner：weichen。
+
+## 基线与授权
+
+- 权威仓：weichen/Agent-System-Core；root_target：main；immediate_target：main
+- branch_kind：initiative；initiative_id：governance-adoption-Agent-System-Core
+- R0（initial_main_sha）：35560aa5a3fdc9d8f92bb4b2e9eece5254b94c41；工作区：/home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption（worktree）；分支：feature/governance-adoption
+- 父任务信息（仅子任务填写）：无
+- accepting_requirements：false（单仓采纳 initiative，范围固定为下列十步，不接收新需求）
+- 需求分流：verdict=NEW；selected_initiative=无；候选=无；理由=development-governance-v2 Phase 6 每个正式仓库独立 initiative；triage 收据=无（本仓采纳前尚无 triage 入口）
+- 授权与交付终点：owner 已授权 development-governance-v2 全部实施（Agent-System-Dev 主记录 docs/superpowers/plans/2026-10-16-development-governance-v2-master.md）；终点为本 PR 合入 main 且采纳状态 active。分支保护与 tag 由用户执行。
+- 实施执行：agent-per-unit（配置：AGENT-EXECUTION.md；外部 Agent 一单元一新进程/会话/工作区）
+
+## 目标与非目标
+
+目标（Phase 6 十步）：固定 Guidelines SHA（deps/development-guidelines @ 2688fd2be6a52c0bf110e7e6366b6f291ab5e4eb）；增加 AGENT-EXECUTION.md；增加共享执行器薄入口（deps/development-manager @ 42e675fec8e0069646319fcd298e61d03d8cbfd5 + scripts/dm、scripts/agent-exec、scripts/task）；四条路由（implementation/repair/test/review）；任务卡模板与分支元数据；CI 中启用 schema 校验（.gitea/workflows/governance.yml：agent-exec validate + governance check）；一次正例运行；负例（缺配置、坏收据、越界路径、未知副作用）；独立 Review + CI；采纳状态 active。
+
+非目标：不改业务代码与既有测试；不改分支保护（由用户执行）；不发布 tag。
+
+引导说明：AGENT-EXECUTION.md 由 host 以固定版本治理套件（dm-server governance init @ 42e675fec8e0069646319fcd298e61d03d8cbfd5）生成并作为引导配置提交（执行器运行前必须存在有效配置；沿用 DM 采纳先例）；其余文件由实施单元生成。
+
+## Agent 实施收据
+
+| 单元 ID | 预期范围 | 角色 | 候选顺序 | attempt | CLI/模型 | 终态 | transport | fallback | 源码外收据路径 |
+|---|---|---|---|---|---|---|---|---|---|
+| governance-adoption-Agent-System-Core-r1 | 13 个允许路径（治理文件 + 两个 gitlink） | dde4f359 | implementation | cursor→antigravity→pi | run 9a7ec135（dm-server 42e675f，attempts=1） | cursor / not-observable | success | 98cfc6408d49（apply 快进） | 否 | /tmp/gov-receipt-Agent-System-Core-r1.json |
+| governance-adoption-Agent-System-Core-repin-r1 | 9 个允许路径（治理文件 + 两个 gitlink） | a160b03c | implementation | cursor→antigravity→pi | run e64b5afe（dm-server pinned，attempts=1） | cursor / not-observable | success | 1ada9955d57a（apply 快进） | 否 | /tmp/gov-receipt-Agent-System-Core-repin-r1.json |
+
+## 同步记录
+
+| 节点 | sync_base_sha | tested_head_sha | checked_at | 结果 |
+|---|---|---|---|---|
+| 启动 | 35560aa5a3fdc9d8f92bb4b2e9eece5254b94c41 | 35560aa5a3fdc9d8f92bb4b2e9eece5254b94c41 | 2026-10-17T05:51:28Z | 一致 |
+
+## 结果与验证（完成后填写）
+
+- 结果：治理套件已采纳——Guidelines 与 Development-Manager gitlink 固定、AGENT-EXECUTION.md 四条路由、共享执行器薄入口、pre-commit admit、governance CI；采纳状态 active（随合入生效）。
+- 验证：见下表（命令输出由 /tmp/gov-neg/ev.sh 写入 .task/evidence/governance-adoption/，SELF 除非另注）。
+- 影响与剩余项：不改业务代码；本地启用钩子需由人执行 `git config core.hooksPath scripts/hooks`；将 governance 检查设为分支保护必需项属用户双钥匙事项（命令见 Agent-System-Dev 治理主记录）。
+
+| 时间(UTC) | 命令 | 被测 HEAD | exit | 来源 | 摘要/位置 |
+|---|---|---|---|---|---|
+| 20261017T061900Z | `scripts/dm governance check --repo .` | 98cfc6408d49 | 0 | SELF | ok fixtures；[输出](../evidence/governance-adoption/00-20261017T061900Z.log) sha256:ae4439e39cad7e8d |
+| 20261017T061900Z | `scripts/agent-exec validate --repo .` | 98cfc6408d49 | 0 | SELF | validate: ok；[输出](../evidence/governance-adoption/01-20261017T061900Z.log) sha256:89ea92d2c3d1f9e4 |
+| 20261017T061900Z | `python3 /tmp/gov-neg/adopt_cases.py .` | 98cfc6408d49 | 0 | SELF | [verdict] ALL PASS cases=5；[输出](../evidence/governance-adoption/02-20261017T061900Z.log) sha256:64f13535108476f9 |
+| 20261017T071536Z | `scripts/dm governance check --repo .` | 1ada9955d57a | 0 | SELF | ok fixtures；[输出](../evidence/governance-adoption/03-20261017T071536Z.log) sha256:f46b6a39dd69f693 |
+| 20261017T071536Z | `scripts/agent-exec validate --repo .` | 1ada9955d57a | 0 | SELF | validate: ok；[输出](../evidence/governance-adoption/04-20261017T071536Z.log) sha256:4a0c76e79c70e3bb |
+| 20261017T071536Z | `python3 /tmp/gov-neg/adopt_cases.py .` | 1ada9955d57a | 0 | SELF | [verdict] ALL PASS cases=5；[输出](../evidence/governance-adoption/05-20261017T071536Z.log) sha256:3eb026a392f8336b |
+| — | 候选 HEAD CI | 18f617d4e8cd | 0 | GATE | success governance / governance (pull_request)=success（/weichen/Agent-System-Core/actions/runs/3535/jobs/5611） |
+| — | 独立 Review | 18f617d4e8cd | 0 | REPLAY | cursor-agent-stable --mode ask 全新只读会话（fresh_session=true，session id not-observable；codex/pi 因 Token Plan 429 不可用）VERDICT: APPROVE，无阻塞项 |
@@ -0,0 +1,19 @@
+command: scripts/dm governance check --repo . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 98cfc6408d49dd88d957bb2dd13ca37bebe9facf
+exit: 0
+utc: 20261017T061900Z
+---
+ok gitlink deps/development-guidelines
+ok gitlink deps/development-manager
+ok agent-execution
+ok file AGENTS.md
+ok file PROJECT.md
+ok file policy/guidelines-adoption.md
+ok file .task/templates/task-card.md
+ok executable scripts/agent-exec
+ok executable scripts/hooks/pre-commit
+ok executable scripts/dm
+ok executable scripts/task
+ok adoption status
+ok fixtures
@@ -0,0 +1,7 @@
+command: scripts/agent-exec validate --repo . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 98cfc6408d49dd88d957bb2dd13ca37bebe9facf
+exit: 0
+utc: 20261017T061900Z
+---
+validate: ok
@@ -0,0 +1,13 @@
+command: python3 /tmp/gov-neg/adopt_cases.py . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 98cfc6408d49dd88d957bb2dd13ca37bebe9facf
+exit: 0
+utc: 20261017T061900Z
+---
+[object] Agent-System-Core-wt-feature-governance-adoption HEAD=98cfc6408d49dd88d957bb2dd13ca37bebe9facf
+PASS positive-run-apply: run exit=0 terminal=success apply exit=0 head=b38a8a0da1df
+PASS missing-config: run exit=2 terminal=invalid-config validate exit=2
+PASS bad-receipt: apply exit=4 head_unchanged=True msg=apply: no local integrity record: changed paths do not match the transport diff
+PASS out-of-scope-path: run exit=5 terminal=scope-expansion transport=False
+PASS unknown-side-effect: run exit=4 terminal=failed-unknown transport=False
+[verdict] ALL PASS cases=5
@@ -0,0 +1,19 @@
+command: scripts/dm governance check --repo . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 1ada9955d57af1dcfb750db2f01a1bcc40fd3ac4
+exit: 0
+utc: 20261017T071536Z
+---
+ok gitlink deps/development-guidelines
+ok gitlink deps/development-manager
+ok agent-execution
+ok file AGENTS.md
+ok file PROJECT.md
+ok file policy/guidelines-adoption.md
+ok file .task/templates/task-card.md
+ok executable scripts/agent-exec
+ok executable scripts/hooks/pre-commit
+ok executable scripts/dm
+ok executable scripts/task
+ok adoption status
+ok fixtures
@@ -0,0 +1,7 @@
+command: scripts/agent-exec validate --repo . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 1ada9955d57af1dcfb750db2f01a1bcc40fd3ac4
+exit: 0
+utc: 20261017T071536Z
+---
+validate: ok
@@ -0,0 +1,13 @@
+command: python3 /tmp/gov-neg/adopt_cases.py . 
+cwd: /home/weichen/projects/Agent-System-Core-wt-feature-governance-adoption
+head: 1ada9955d57af1dcfb750db2f01a1bcc40fd3ac4
+exit: 0
+utc: 20261017T071536Z
+---
+[object] Agent-System-Core-wt-feature-governance-adoption HEAD=1ada9955d57af1dcfb750db2f01a1bcc40fd3ac4
+PASS positive-run-apply: run exit=0 terminal=success apply exit=0 head=f6303ea75059
+PASS missing-config: run exit=2 terminal=invalid-config validate exit=2
+PASS bad-receipt: apply exit=4 head_unchanged=True msg=apply: no local integrity record: changed paths do not match the transport diff
+PASS out-of-scope-path: run exit=5 terminal=scope-expansion transport=False
+PASS unknown-side-effect: run exit=4 terminal=failed-unknown transport=False
+[verdict] ALL PASS cases=5
@@ -0,0 +1,66 @@
+# <任务标题> 任务卡
+
+任务主记录：<Issue/PR 链接，或注明"本文件随 PR 描述维护">。
+状态：<进行中 / 待审核 / 已合入 / 已放弃>；owner：<填写>。
+
+## 基线与授权
+
+- 权威仓：<仓 URL 或逻辑 ID>；root_target：main；immediate_target：<main 或 feature/initiative>
+- branch_kind：<initiative | child | standalone>；initiative_id：<稳定 ID，普通任务为空>
+- R0（initial_main_sha）：<完整 SHA>；工作区：<路径>（worktree / clone）；分支：<feature|fix|hotfix/...>
+- 父任务信息（仅子任务填写）：parent_branch=<父分支>；parent_head_sha=<SHA>；parent_pr=<父 PR>
+- accepting_requirements：<true | false>（仅 initiative 可为 true；冻结后为 false）
+- 需求分流：verdict=<JOIN | NEW>；selected_initiative=<分支或无>；候选=<分支@HEAD 列表或无>；理由=<不适配/适配依据>；triage 收据=<源码外路径与摘要>
+- 授权与交付终点：<本次允许的动作与终点；未授权事项写明>
+- 实施执行：agent-per-unit（配置：AGENT-EXECUTION.md；外部 Agent 一单元一新进程/会话/工作区）
+
+## 目标与非目标
+
+<预期结果、受影响入口/模块/契约；明确不做什么。>
+
+## Agent 实施收据
+
+按根目录 AGENT-EXECUTION.md 路由；外部 Agent 一单元一新进程/会话/隔离工作区。实施、修复与测试必须通过登记的候选执行；未配置或候选安全耗尽停在 unavailable/blocked，禁止 host 隐式代写。
+
+| 单元 ID | 预期范围 | 角色 | 候选顺序 | attempt | CLI/模型 | 终态 | transport | fallback | 源码外收据路径 |
+|---|---|---|---|---|---|---|---|---|---|
+| 待定义 | — | NOT_RUN | NOT_RUN | NOT_RUN | NOT_RUN | NOT_RUN | NOT_RUN | NOT_RUN | — |
+
+## 复用材料（task-start 自动检索结果摘录）
+
+- 相关近期提交：<git log -- <paths> 命中；无则写无>。
+- 相关经验条目：<检索命中；无则写无>。
+
+## 同步记录
+
+| 节点 | sync_base_sha | tested_head_sha | checked_at | 结果 |
+|---|---|---|---|---|
+| 启动 | <R0> | <HEAD> | <日期时间> | <一致/已同步> |
+| 验收前 | <填写> | <填写> | <填写> | <填写> |
+
+字段没有发生的写"无"，不省略。本卡与 PR 描述同源：开 PR 时将"目标与非目标、结果与验证、影响与剩余项"复制进 PR 描述，不另写第二份。
+
+## 交付调度（有多个就绪任务时）
+
+- 调度记录 / entry_id / 当前 attempt_id：<平台或调度服务链接及标识；未入队写无>。
+- 入队源 HEAD / owner / 依赖：<完整 SHA、身份、依赖或无>。
+- 固定候选引用：<调度尝试中的 base/head、组件与配置；未固定写无>。
+- 窗口、最终 CI、Review 与真实合入结果：<引用调度记录及原始证据，不复制全队列>。
+
+定义见 [合入调度与 CI 分配](../standards/merge-queue.md)。本段在候选固定前整理；固定后先使用源码之外的记录保存运行结果，再按项目闭合流程回填。
+
+## 结果与验证（完成后填写）
+
+- 结果：<交付了什么；未发生写无>。
+- 验证：<结论性摘要：PASS/FAIL/SKIP/NOT_RUN 逐项；逐条证据见下表，来源标注与写入规则见 [按影响选择验证与证据](../standards/verification.md)>。
+- 影响与剩余项：<影响面、遗留风险、后续任务>。
+
+| 时间(UTC) | 命令 | 被测 HEAD | exit | 来源 | 摘要/位置 |
+|---|---|---|---|---|---|
+| <时间> | <命令；由项目 evidence 工具写入，无工具时附命令+对象 SHA+输出 digest 并标 SELF> | <HEAD> | <exit> | GATE/REPLAY/SELF | <摘要/位置> |
+
+## 回收（task-finish）
+
+- 合入核对：<分支已包含于目标分支的核对输出>。
+- 未跟踪/ignored 文件处置：<归档/可再生/存疑 逐项；无则写无>。
+- 回收动作：<worktree remove、分支删除、prune；放弃路径注明原因>。
@@ -0,0 +1,67 @@
+# Project Agent Execution Configuration
+# 项目工程 Agent 执行路由配置
+
+适用范围：定义本项目中由 AI / Agent 主导的代码实施、修复、测试与审阅所调用的 Agent CLI 路由。
+解析器识别正文中唯一一个 ```agent-execution 代码块。正文外的说明文字可根据项目需要调整。
+
+通用硬规则见 Development-Guidelines 的 [Agent 执行与协作契约](deps/development-guidelines/standards/agent-execution.md)。
+本配置为项目本地路由权威，声明具体的候选 profile、尝试次数上限、超时、集成入口与豁免路径；
+代码块须通过 `deps/development-guidelines/templates/schemas/agent-execution.schema.json` 校验后才能登记为 active。
+`fallback_to_host` 只能为 `never`；旧值 `all-unavailable` 仅供迁移工具读取，不能通过准入。
+下例的 CLI 名称、命令与路径均为占位，按项目实际执行器替换。
+
+```agent-execution
+schema: agent-execution/v1
+
+execution:
+  fresh_process_per_unit: true
+  permission: run-everything
+  fallback_to_host: never
+  max_parallel: 3
+  timeout_seconds: 7200
+  idle_timeout_seconds: 600
+  summary_bytes: 16384
+  raw_output_bytes: 10485760
+  max_attempts_per_profile: 3
+  emergency_profile: codex
+
+roles:
+  implementation: [cursor, antigravity, pi]
+  repair: [cursor, antigravity, pi]
+  test: [pi, antigravity, cursor]
+  review: [pi, antigravity, cursor, codex]
+
+profiles:
+  cursor:
+    adapter: cursor
+    executable: "cursor-agent-stable"
+    model: auto
+    enabled: true
+
+  antigravity:
+    adapter: antigravity
+    executable: agy
+    model: auto
+    enabled: true
+
+  pi:
+    adapter: pi
+    executable: pi
+    model: auto
+    enabled: true
+
+  codex:
+    adapter: codex
+    executable: codex
+    model: auto
+    enabled: true
+
+integration:
+  exempt_paths: [".task/**", "docs/superpowers/specs/**", "docs/superpowers/plans/**"]
+  transport:
+    command: "scripts/agent-exec apply"
+  admission:
+    command: "scripts/agent-exec admit"
+  cleanup:
+    command: "scripts/agent-exec cleanup"
+```
@@ -34,3 +34,21 @@
 - **SSH key**：`~/.ssh/home_gitea`（ED25519，公钥指纹
  `SHA256:GYXaOXIMeloQ1fKSfdJ/ndgPQzs/Ix7Hk62ekfkfQ20`，Gitea 侧密钥名 GMKtec）；
  `~/.ssh/config` 已有 `git-ssh.u1680832.nyat.app` 主机条目，克隆/推送直接可用。
+
+## Agent 执行与新项目准入（强制）
+
+1. 缺 Guidelines 固定版本（`deps/development-guidelines` gitlink 未就绪或 HEAD 不符）时，Agent 只能只读分析，不得实施。
+2. 缺有效 `AGENT-EXECUTION.md`（`scripts/dm governance check` 未通过）时禁止 AI 实施。
+3. 缺共享执行器（`scripts/agent-exec` 或 `deps/development-manager` 不可用）时，状态为 `blocked/unavailable`，不得改由 host 替代。
+4. 不允许 host 代写实施单元；实施、修复与测试必须经 `AGENT-EXECUTION.md` 登记的候选执行。
+5. 人类直接开发按其自身授权与项目规则进行，不受本条款约束。
+6. 分支保护与规范 tag 由用户负责设置与维护，Agent 不得代为设置。
+
+AI 实施单元只能通过 `scripts/agent-exec run` 运行（配置与路由见 AGENT-EXECUTION.md），并由 `scripts/agent-exec apply` 集成。
+
+入口：
+
+- 需求分流：`scripts/task triage discover|decide|verify`
+- 任务启动：`scripts/task start …`
+- PR 检查：`scripts/task pr-check …`
+- Agent 执行：`scripts/agent-exec run|apply|admit|validate`
@@ -8,6 +8,9 @@
 在本仓根目录执行 `go test ./...` 验证 Go 契约与不变量；纯文档修改执行
 `git diff --check` 并核对受影响链接、命令和来源。本仓没有 Makefile。

+治理门禁（与 `.gitea/workflows/governance.yml` 相同）：在本仓根目录依次执行
+`scripts/agent-exec validate --repo .` 与 `scripts/dm governance check --repo .`，两者均须通过。
+
 修改 `ts/types.ts` 后在 Dev 仓运行 `make contracts` 同步消费者，再以 `make test`
 验证固定组件组合；单独运行 Core 测试不能代替提供方与消费者的兼容验证。
 本仓提供公共契约，不是独立部署服务；当前能力以源码和对应验证记录为准。
@@ -12,20 +12,21 @@ import "time"
 type ActivityCard struct {
 	ID ID `json:"id"`
 	// ProjectID/ConversationID 定位归属；ConversationID 为空表示项目级卡片（远期跨会话收件箱）。
-	ProjectID       ID         `json:"projectId"`
-	ConversationID  ID         `json:"conversationId,omitempty"`
-	SourceRunID     ID         `json:"sourceRunId,omitempty"`     // 溯源：产生卡片的 run
-	AnchorMessageID ID         `json:"anchorMessageId,omitempty"` // 互定位：出生消息（↙ 定位到对话）
-	Kind            CardKind   `json:"kind"`
-	Title           string     `json:"title"`
-	Status          CardStatus `json:"status"`
-	Tone            string     `json:"tone,omitempty"` // ok | warn | danger | info
-	Body            []UINode   `json:"body,omitempty"`
-	Actions         []UIAction `json:"actions,omitempty"` // 走命令总线；未绑定 command 降级为发消息
-	Pinned          bool       `json:"pinned,omitempty"`
-	Work            *WorkCard  `json:"work,omitempty"` // optional background-work snapshot; never a foreground approval
-	CreatedAt       time.Time  `json:"createdAt"`
-	UpdatedAt       time.Time  `json:"updatedAt"`
+	ProjectID       ID            `json:"projectId"`
+	ConversationID  ID            `json:"conversationId,omitempty"`
+	SourceRunID     ID            `json:"sourceRunId,omitempty"`     // 溯源：产生卡片的 run
+	AnchorMessageID ID            `json:"anchorMessageId,omitempty"` // 互定位：出生消息（↙ 定位到对话）
+	Kind            CardKind      `json:"kind"`
+	Title           string        `json:"title"`
+	Status          CardStatus    `json:"status"`
+	Tone            string        `json:"tone,omitempty"` // ok | warn | danger | info
+	Task            *TaskSnapshot `json:"task,omitempty"`
+	Body            []UINode      `json:"body,omitempty"`
+	Actions         []UIAction    `json:"actions,omitempty"` // 走命令总线；未绑定 command 降级为发消息
+	Pinned          bool          `json:"pinned,omitempty"`
+	Work            *WorkCard     `json:"work,omitempty"` // optional background-work snapshot; never a foreground approval
+	CreatedAt       time.Time     `json:"createdAt"`
+	UpdatedAt       time.Time     `json:"updatedAt"`
 }

 // WorkCard keeps the complete saved result and an optional, non-blocking next step
@@ -13,18 +13,19 @@ type ParsedDocument struct {
 }

 type DocMeta struct {
-	Filename   string    `json:"filename"`
-	MIME       string    `json:"mime,omitempty"`
-	Parser     string    `json:"parser"`              // builtin:md / builtin:docx / mineru / agent
-	Confidence float64   `json:"confidence,omitempty"` // 0–1；交叉验证差异块拉低
-	CrossCheck *CrossCheck `json:"crossCheck,omitempty"`
-	ParsedAt   time.Time `json:"parsedAt"`
+	Origin     *DocumentOrigin `json:"origin,omitempty"`
+	Filename   string          `json:"filename"`
+	MIME       string          `json:"mime,omitempty"`
+	Parser     string          `json:"parser"`               // builtin:md / builtin:docx / mineru / agent
+	Confidence float64         `json:"confidence,omitempty"` // 0–1；交叉验证差异块拉低
+	CrossCheck *CrossCheck     `json:"crossCheck,omitempty"`
+	ParsedAt   time.Time       `json:"parsedAt"`
 }

 type CrossCheck struct {
-	With     string  `json:"with"` // 交叉验证的另一解析器
-	DiffBlocks int   `json:"diffBlocks"`
-	Verdict  string  `json:"verdict"` // agree | differ
+	With       string `json:"with"` // 交叉验证的另一解析器
+	DiffBlocks int    `json:"diffBlocks"`
+	Verdict    string `json:"verdict"` // agree | differ
 }

 type DocBlockType string
@@ -39,13 +40,13 @@ const (
 )

 type DocBlock struct {
-	Type    DocBlockType `json:"type"`
-	Level   int          `json:"level,omitempty"` // heading
-	Text    string       `json:"text,omitempty"`  // para/heading/list 项拼接/code
-	Items   []string     `json:"items,omitempty"` // list
-	Rows    [][]string   `json:"rows,omitempty"`  // table（首行为表头）
-	Ref     string       `json:"ref,omitempty"`   // image-ref 的对象键
-	Confidence float64   `json:"confidence,omitempty"`
+	Type       DocBlockType `json:"type"`
+	Level      int          `json:"level,omitempty"` // heading
+	Text       string       `json:"text,omitempty"`  // para/heading/list 项拼接/code
+	Items      []string     `json:"items,omitempty"` // list
+	Rows       [][]string   `json:"rows,omitempty"`  // table（首行为表头）
+	Ref        string       `json:"ref,omitempty"`   // image-ref 的对象键
+	Confidence float64      `json:"confidence,omitempty"`
 }

 // PlainText 把文档压成纯文本（附件注入与全文索引用）；表格按行拼接。
@@ -88,60 +89,60 @@ func (d ParsedDocument) PlainText() string {

 // ParserDesc 注册表条目（来源标记：builtin/plugin/...；一切可禁用）。
 type ParserDesc struct {
-	ID      string    `json:"id"`
-	Formats []string  `json:"formats"` // 支持的扩展名
+	ID      string     `json:"id"`
+	Formats []string   `json:"formats"` // 支持的扩展名
 	Source  SourceKind `json:"source"`
-	Note    string    `json:"note,omitempty"`
+	Note    string     `json:"note,omitempty"`
 }

 // ---- 检索（4.12）----

 // KnowledgeDoc 是检索语料的一条文档（来自 ingest 或插件知识库，带来源标记）。
 type KnowledgeDoc struct {
-	ID       ID        `json:"id"`
-	Owner    ID        `json:"owner,omitempty"`     // 空 = 共享（插件级）知识
-	ProjectID ID       `json:"projectId,omitempty"` // 空 = 插件级知识库
-	Source   string    `json:"source"`              // upload | artifact | plugin:<id>
-	Title    string    `json:"title"`
-	Ref      string    `json:"ref,omitempty"`       // 工件/文件引用
-	Chunk    string    `json:"chunk"`
+	ID        ID        `json:"id"`
+	Owner     ID        `json:"owner,omitempty"`     // 空 = 共享（插件级）知识
+	ProjectID ID        `json:"projectId,omitempty"` // 空 = 插件级知识库
+	Source    string    `json:"source"`              // upload | artifact | plugin:<id>
+	Title     string    `json:"title"`
+	Ref       string    `json:"ref,omitempty"` // 工件/文件引用
+	Chunk     string    `json:"chunk"`
 	CreatedAt time.Time `json:"createdAt"`
 }

 type RetrievalHit struct {
-	Doc    KnowledgeDoc `json:"doc"`
-	Score  float64      `json:"score"`
-	Kind   string       `json:"kind"` // fulltext | vector | plugin
+	Doc   KnowledgeDoc `json:"doc"`
+	Score float64      `json:"score"`
+	Kind  string       `json:"kind"` // fulltext | vector | plugin
 }

 type RetrievalQuery struct {
-	ProjectID ID     `json:"projectId,omitempty"`
-	Text      string `json:"text"`
-	TopK      int    `json:"topK,omitempty"`
+	ProjectID ID       `json:"projectId,omitempty"`
+	Text      string   `json:"text"`
+	TopK      int      `json:"topK,omitempty"`
 	Sources   []string `json:"sources,omitempty"` // 过滤来源
 }

 // ---- 记忆（4.13 契约；实现远期）----

 type MemoryEntry struct {
-	ID          string    `json:"id"`
-	Scope       string    `json:"scope"`  // user | project
-	ProjectID   ID        `json:"projectId,omitempty"`
-	Type        string    `json:"type"`   // fact | preference | decision | episode | procedure
-	Content     string    `json:"content"`
-	Source      string    `json:"source"` // run:<id> | manual
-	Confidence  float64   `json:"confidence,omitempty"`
-	CreatedAt   time.Time `json:"createdAt"`
-	LastAccessed time.Time `json:"lastAccessed"`
-	ExpiresAt   *time.Time `json:"expiresAt,omitempty"`
+	ID           string     `json:"id"`
+	Scope        string     `json:"scope"` // user | project
+	ProjectID    ID         `json:"projectId,omitempty"`
+	Type         string     `json:"type"` // fact | preference | decision | episode | procedure
+	Content      string     `json:"content"`
+	Source       string     `json:"source"` // run:<id> | manual
+	Confidence   float64    `json:"confidence,omitempty"`
+	CreatedAt    time.Time  `json:"createdAt"`
+	LastAccessed time.Time  `json:"lastAccessed"`
+	ExpiresAt    *time.Time `json:"expiresAt,omitempty"`
 }

 // ---- 上传文件 ----

 type UploadInfo struct {
-	ArtifactID ID     `json:"artifactId"`
-	Filename   string `json:"filename"`
-	MIME       string `json:"mime,omitempty"`
-	Size       int64  `json:"size"`
+	ArtifactID ID              `json:"artifactId"`
+	Filename   string          `json:"filename"`
+	MIME       string          `json:"mime,omitempty"`
+	Size       int64           `json:"size"`
 	Parsed     *ParsedDocument `json:"parsed,omitempty"`
 }
@@ -0,0 +1,37 @@
+# 通用规范采纳记录
+
+- 项目/负责人：Agent-System-Core / weichen
+- 状态：active
+- 规范位置：`deps/development-guidelines`
+- 规范性文档目录：`policy/`（本仓现有 `policy/README.md`；项目规则与制度放此处）
+- 过程文档位置：仓内固定目录 `.task/`（任务卡与任务主记录；任务主记录同时随 PR 描述维护，豁免理由见「项目差异」）
+- 版本权威：本仓 gitlink `deps/development-guidelines` @ `2688fd2be6a5`（完整 SHA `2688fd2be6a52c0bf110e7e6366b6f291ab5e4eb`）；可读标签：无
+- 批准/生效依据：feature/governance-adoption PR（Agent-System-Core）
+- 替代范围和迁移说明：无
+
+| 规范/工作流 | 适用条件与范围 | 采纳或不适用理由 | 执行方式/证据 |
+|---|---|---|---|
+| core/baseline.md | 日常工作 | 采纳：所有任务的硬边界、默认做法与完成声明基线 | manual |
+| workflows/delivery.md、debugging.md、handoff.md | 交付、排障与交接 | 采纳：本仓以 PR 交付，Bug 排障与会话交接走统一流程 | manual |
+| standards/collaboration.md | 并行共同起点、两级分支、节点同步与陈旧候选拒绝 | 采纳：PR 合入前须同步最新目标分支并附证据（见 AGENTS.md） | manual；暂无真实平台负例，未登记 enforced |
+| standards/merge-queue.md | 多个就绪任务的最终同步、CI、Review 与合入调度 | 不适用：本仓由单一 owner 串行合入，当前无多个就绪任务竞争交付窗口；出现时另行登记 | 不适用 |
+| standards/agent-execution.md | 工程 Agent CLI 执行路由、会话隔离与受控集成 | 采纳：允许 AI 协助代码实施；配置见 AGENT-EXECUTION.md | AGENT-EXECUTION.md + scripts/agent-exec（validate / admit）；证据：governance check 与正负例（.task/feature-governance-adoption.md） |
+| standards/verification.md | 验证与结论 | 采纳：完成声明须附实际命令输出 | 命令见 PROJECT.md「验证」节（`go test ./...`、`git diff --check`、`scripts/agent-exec validate --repo .`、`scripts/dm governance check --repo .`）；CI：`.gitea/workflows/governance.yml` 运行治理门禁 |
+| standards/module-contracts.md、standards/multi-repo.md | 公共契约（Go 类型、黄金样本、TS 镜像）的边界，以及依赖仓以 gitlink 固定版本 | 采纳：本仓是公共契约库，治理依赖以 gitlink 固定 | manual；gitlink 与 `.gitmodules` 固定，CI 物化白名单见 `.gitea/workflows/governance.yml`；契约一致性由 `conformance_test.go` 与 `testdata/golden` 覆盖 |
+| standards/security.md | CI 凭据、submodule 取回与执行工具边界 | 采纳：CI 仅以限定 URL 的凭据取回两个治理 gitlink，凭据不落盘 | manual；`.gitea/workflows/governance.yml` 白名单与凭据约束 |
+| workflows/design-verification.md、standards/testability.md | 契约增量的设计与验证、纯函数的可独立测试性 | 采纳：契约变更按增量设计验收，并以黄金样本与一致性测试验证 | manual；`conformance_test.go`、`testdata/golden` |
+| standards/failure-containment.md、data-reliability.md、observability.md、configuration.md、release-operations.md、automation.md | 运行时服务、数据库、部署发布、定时自动化与故障控制 | 不适用：本仓只含契约与纯函数，不含 I/O、数据库或运行服务，不独立部署（见 PROJECT.md） | 不适用 |
+
+## 项目差异
+
+| 原规则/章节 | 项目采用方式 | 原因与风险 | 负责人/复查条件 |
+|---|---|---|---|
+| 过程文档默认放独立过程仓 | 任务卡与任务主记录放仓内 `.task/`，AGENT-EXECUTION.md 的 `exempt_paths` 已豁免 `.task/**` | 单一 owner 的小仓库，过程记录体量小，独立过程仓成本不划算；风险是过程记录与代码同仓，禁止写入凭据或 token | weichen；协作人增加或过程记录需跨仓共享时复查 |
+
+## 本次升级或首次接入验证
+
+- 入口：`AGENTS.md` 保留原内容并追加「Agent 执行与新项目准入（强制）」；`PROJECT.md`「验证」节补充 `scripts/agent-exec validate --repo .` 与 `scripts/dm governance check --repo .`；`AGENT-EXECUTION.md` 原样保留（`governance init --adopt` 跳过已存在文件）。
+- 版本获取：`deps/development-guidelines` @ `2688fd2be6a5`，`deps/development-manager` @ `bd3a7de57a3a`；`.gitmodules` 使用规范 SSH 地址，不含凭证或本机路径。
+- 占位符：模板占位符已全部替换；`policy/guidelines-adoption.md` 与追加的准入段中不再有模板尖括号文字。
+- 冲突检查：仓内原无其他采纳记录（无 `policy/adoption.md`），原无旧版 Agent 执行入口（无 `scripts/agent_exec.py`），原无 `.gitea/workflows`，治理 workflow 为新增，无需修改既有 CI。
+- 正负例与 Review/CI 证据：见 `.task/feature-governance-adoption.md`。
@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+# 共享 Agent 执行器薄入口：转调 scripts/dm 的 agent-exec 子命令（validate / run / apply / admit / cleanup）。
+set -euo pipefail
+exec "$(dirname "$0")/dm" agent-exec "$@"
@@ -0,0 +1,55 @@
+#!/usr/bin/env bash
+# 共享 dm-server 薄入口：读取 deps/development-manager 的 gitlink 钉住提交，
+# 按该提交构建并缓存 dm-server，再 exec 其子命令（agent-exec、task、governance）。执行语义与退出码以该二进制为准。
+set -euo pipefail
+
+root=$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd -P)
+dm="$root/deps/development-manager"
+
+die() { echo "dm: $*" >&2; exit 2; }
+
+# git 钩子会导出仓库本地变量（GIT_DIR、GIT_INDEX_FILE 等），它们会让 `git -C "$dm"` 解析到外层仓库。
+# 所有针对 DM 检出的 git 与 go 调用都在清除这些变量后执行；读取外层 gitlink 与最终 exec 仍保留继承环境。
+dm_local_env="GIT_DIR GIT_WORK_TREE GIT_INDEX_FILE GIT_PREFIX GIT_OBJECT_DIRECTORY GIT_ALTERNATE_OBJECT_DIRECTORIES GIT_COMMON_DIR GIT_GRAFT_FILE GIT_NAMESPACE GIT_CONFIG GIT_CONFIG_PARAMETERS GIT_CONFIG_COUNT GIT_SHALLOW_FILE GIT_IMPLICIT_WORK_TREE $(git rev-parse --local-env-vars 2>/dev/null || true)"
+dm_clean_env() { for v in $dm_local_env; do unset "$v"; done; }
+dmgit() ( dm_clean_env; git -C "$dm" "$@" )
+
+# 1. 钉住版本只来自索引中的 gitlink（mode 160000），不读取工作区文件。
+entry=$(git -C "$root" ls-files -s -- deps/development-manager 2>/dev/null || true)
+[ -n "$entry" ] || die "索引中缺少 deps/development-manager 的 gitlink，无法确定钉住的 DM 版本"
+read -r mode pin _stage _path <<<"$entry"
+[ "$mode" = 160000 ] || die "deps/development-manager 不是 gitlink（mode=$mode）"
+[[ "$pin" =~ ^[0-9a-f]{40}$ ]] || die "gitlink 提交格式无效：$pin"
+
+# 2. 检出必须存在且确为 DM 独立仓库（避免 git -C 误解析到外层仓库），HEAD 必须等于钉住提交。
+[ -e "$dm/.git" ] || die "deps/development-manager 未检出；请按钉住提交 $pin 检出该 gitlink"
+top=$(dmgit rev-parse --show-toplevel 2>/dev/null || true)
+[ -n "$top" ] && [ "$top" = "$(cd "$dm" && pwd -P)" ] \
+  || die "deps/development-manager 不是独立检出；请按钉住提交 $pin 检出该 gitlink"
+head=$(dmgit rev-parse HEAD 2>/dev/null || true)
+[ "$head" = "$pin" ] \
+  || die "deps/development-manager HEAD（${head:-未知}）与钉住提交 $pin 不一致；请按钉住提交检出"
+
+# 3. 检出工作树必须干净（忽略嵌套 submodule 状态）。
+dirty=$(dmgit status --porcelain --ignore-submodules=all) \
+  || die "无法读取 deps/development-manager 的工作树状态"
+[ -z "$dirty" ] || die "deps/development-manager 存在未提交改动，拒绝运行（请先清理该检出）"
+
+# 4. 二进制按钉住提交缓存；缺失时在缓存目录内临时构建，完成后原子改名，避免留下半成品。
+cache="${DM_CACHE:-${XDG_CACHE_HOME:-$HOME/.cache}/development-manager/dm-server}"
+bin="$cache/dm-server-$pin"
+if [ ! -x "$bin" ]; then
+  command -v go >/dev/null 2>&1 || die "缺少 go，无法构建 dm-server（$pin）"
+  mkdir -p "$cache"
+  cache=$(cd "$cache" && pwd -P)
+  bin="$cache/dm-server-$pin"
+  work=$(mktemp -d "$cache/build-$pin.XXXXXX")
+  if ! (dm_clean_env; cd "$dm" && GOWORK=off go build -trimpath -o "$work/dm-server" ./cmd/dm-server); then
+    rm -rf "$work"
+    die "构建 dm-server 失败（$pin）"
+  fi
+  mv -f "$work/dm-server" "$bin"
+  rm -rf "$work"
+fi
+
+exec "$bin" "$@"
@@ -0,0 +1,16 @@
+#!/usr/bin/env bash
+# 本地检查（pre-commit）：集成分支直提拒绝 + 分支命名 + 本地完整性核对。权威准入在 PR（baseline.md:26）。
+branch=$(git rev-parse --abbrev-ref HEAD)
+case "$branch" in
+  main|master)
+    echo "pre-commit: 拒绝——集成分支 ${branch} 不得直接提交，走 feature 分支 + PR（collaboration.md）" >&2
+    exit 1 ;;
+esac
+echo "$branch" | grep -qE '^(feature|fix|hotfix|release)/[A-Za-z0-9._-]+$' || {
+  echo "pre-commit: 拒绝——分支名须为 feature|fix|hotfix|release/<slug>，当前：${branch}" >&2
+  exit 1
+}
+root=$(git rev-parse --show-toplevel)
+# admit 只是本地完整性/篡改证据（SELF）：同一 OS 用户可读取本地 HMAC 密钥并绕过本钩子，
+# 它不是 GATE，也不是可信身份。缺失或无效的本地记录仍拒绝；交付信任来自精确 HEAD 的 CI 或独立 REPLAY。
+"$root/scripts/agent-exec" admit --repo "$root"
@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+# 通用任务入口：转调 scripts/dm 的 task 子命令（triage / start / pr-check）。
+set -euo pipefail
+exec "$(dirname "$0")/dm" task "$@"
@@ -0,0 +1,330 @@
+package core
+
+import (
+	"encoding/json"
+	"fmt"
+	"strings"
+	"time"
+)
+
+type DocumentOrigin struct {
+	ArtifactID     ID     `json:"artifactId"`
+	ContentHash    string `json:"contentHash"`
+	ParserRevision string `json:"parserRevision"`
+}
+type TaskPhase struct {
+	ID         string    `json:"id"`
+	Label      string    `json:"label"`
+	ObservedAt time.Time `json:"observedAt,omitempty"`
+}
+type TaskSourceRef struct {
+	ID             string `json:"id"`
+	ArtifactID     string `json:"artifactId"`
+	ContentHash    string `json:"contentHash"`
+	ParserRevision string `json:"parserRevision"`
+	BlockIndex     int    `json:"blockIndex"`
+	Label          string `json:"label"`
+	Quote          string `json:"quote,omitempty"`
+	Availability   string `json:"availability,omitempty"`
+}
+type TaskResultRef struct {
+	ArtifactID   ID     `json:"artifactId,omitempty"`
+	SurfaceID    ID     `json:"surfaceId,omitempty"`
+	Version      string `json:"version,omitempty"`
+	ContentHash  string `json:"contentHash,omitempty"`
+	Availability string `json:"availability"`
+	Name         string `json:"name,omitempty"`
+}
+type TaskFinding struct {
+	ID          string          `json:"id"`
+	Text        string          `json:"text"`
+	Impact      string          `json:"impact,omitempty"`
+	Kind        string          `json:"kind"`
+	Status      string          `json:"status"`
+	Revision    int64           `json:"revision"`
+	Sources     []TaskSourceRef `json:"sources"`
+	Assumptions string          `json:"assumptions,omitempty"`
+	Corrects    string          `json:"corrects,omitempty"`
+}
+type TaskSnapshot struct {
+	SchemaVersion          int               `json:"schemaVersion"`
+	TaskID                 ID                `json:"taskId"`
+	AttemptID              ID                `json:"attemptId"`
+	CardID                 ID                `json:"cardId"`
+	ProjectID              ID                `json:"projectId"`
+	ConversationID         ID                `json:"conversationId"`
+	SourceMessageID        ID                `json:"sourceMessageId,omitempty"`
+	ParentRunID            ID                `json:"parentRunId,omitempty"`
+	RunID                  ID                `json:"runId,omitempty"`
+	OwnerInstanceID        string            `json:"ownerInstanceId"`
+	ExecutionEpoch         int64             `json:"executionEpoch,omitempty"`
+	Revision               int64             `json:"revision"`
+	SourceSeq              int64             `json:"sourceSeq"`
+	StreamScope            string            `json:"streamScope"`
+	InputRevision          string            `json:"inputRevision,omitempty"`
+	Title                  string            `json:"title"`
+	Goal                   string            `json:"goal,omitempty"`
+	ExecutionState         string            `json:"executionState"`
+	OutcomeCompleteness    string            `json:"outcomeCompleteness"`
+	DeliveryState          string            `json:"deliveryState"`
+	Phase                  TaskPhase         `json:"phase"`
+	Focus                  string            `json:"focus,omitempty"`
+	Coverage               string            `json:"coverage,omitempty"`
+	Remaining              string            `json:"remaining,omitempty"`
+	NextStep               string            `json:"nextStep,omitempty"`
+	Conclusion             string            `json:"conclusion,omitempty"`
+	Findings               []TaskFinding     `json:"findings,omitempty"`
+	Limitations            []string          `json:"limitations,omitempty"`
+	Inputs                 []TaskSourceRef   `json:"inputs,omitempty"`
+	Actions                []UIAction        `json:"actions,omitempty"`
+	ResultRefs             []TaskResultRef   `json:"resultRefs,omitempty"`
+	LastActivityAt         time.Time         `json:"lastActivityAt,omitempty"`
+	LastMeaningfulUpdateAt time.Time         `json:"lastMeaningfulUpdateAt,omitempty"`
+	CheckpointKeys         map[string]string `json:"checkpointKeys,omitempty"`
+	CheckpointCount        int               `json:"checkpointCount,omitempty"`
+	CheckpointSeq          int64             `json:"checkpointSeq,omitempty"`
+}
+type TaskSourceQuote struct {
+	SourceID   string `json:"sourceId"`
+	BlockIndex int    `json:"blockIndex"`
+	Quote      string `json:"quote"`
+}
+type WorkCheckpointCandidate struct {
+	CheckpointID string                 `json:"checkpointId"`
+	PhaseID      string                 `json:"phaseId"`
+	Focus        string                 `json:"focus,omitempty"`
+	Remaining    string                 `json:"remaining,omitempty"`
+	NextStep     string                 `json:"nextStep,omitempty"`
+	Findings     []WorkFindingCandidate `json:"findings"`
+}
+type WorkFindingCandidate struct {
+	FindingKey  string               `json:"findingKey"`
+	Text        string               `json:"text"`
+	Impact      string               `json:"impact,omitempty"`
+	Kind        string               `json:"kind"`
+	Assumptions string               `json:"assumptions,omitempty"`
+	Corrects    string               `json:"corrects,omitempty"`
+	Sources     []TaskSourceQuote    `json:"sources"`
+	Evidence    *WorkFindingEvidence `json:"evidence,omitempty"`
+}
+
+// WorkFindingEvidence is a model-supplied candidate, not a persisted proof.
+// Backend checks the quoted source text and computes any resulting statement.
+type WorkEvidencePremise struct {
+	Role        string `json:"role"`
+	SourceIndex int    `json:"sourceIndex"`
+	Value       *int   `json:"value,omitempty"`
+}
+
+type WorkFindingEvidence struct {
+	Mode     string                `json:"mode"`
+	Premises []WorkEvidencePremise `json:"premises"`
+	Result   *int                  `json:"result,omitempty"`
+}
+
+func validateWorkFindingEvidence(e *WorkFindingEvidence, sourceCount int) error {
+	if e == nil {
+		return nil
+	}
+	var required []string
+	switch e.Mode {
+	case "direct_relation":
+		required = []string{"direct_relation"}
+		if e.Result != nil {
+			return fmt.Errorf("checkpoint.evidence: direct relation cannot have result")
+		}
+	case "difference":
+		required = []string{"total", "phase"}
+		if e.Result == nil || *e.Result < 0 || *e.Result > 1000000 {
+			return fmt.Errorf("checkpoint.evidence: difference requires bounded result")
+		}
+	case "unresolved_relation":
+		required = []string{"total", "phase", "dependent"}
+		if e.Result != nil {
+			return fmt.Errorf("checkpoint.evidence: unresolved relation cannot have result")
+		}
+	default:
+		return fmt.Errorf("checkpoint.evidence: unknown mode")
+	}
+	if len(e.Premises) != len(required) {
+		return fmt.Errorf("checkpoint.evidence: wrong premise count")
+	}
+	roles := make(map[string]bool, len(required))
+	for _, premise := range e.Premises {
+		if roles[premise.Role] || premise.SourceIndex < 0 || premise.SourceIndex >= sourceCount {
+			return fmt.Errorf("checkpoint.evidence: duplicate role or invalid source index")
+		}
+		roles[premise.Role] = true
+		if e.Mode == "direct_relation" {
+			if premise.Value != nil {
+				return fmt.Errorf("checkpoint.evidence: direct relation cannot have value")
+			}
+		} else if premise.Value == nil || *premise.Value < 0 || *premise.Value > 1000000 {
+			return fmt.Errorf("checkpoint.evidence: count premise requires bounded value")
+		}
+	}
+	for _, role := range required {
+		if !roles[role] {
+			return fmt.Errorf("checkpoint.evidence: missing premise role")
+		}
+	}
+	return nil
+}
+
+// Validate enforces wire limits; owner/source authorization belongs to Backend.
+func (s TaskSnapshot) Validate() error {
+	if s.SchemaVersion != 1 {
+		return fmt.Errorf("task.schema: unsupported")
+	}
+	if s.TaskID == "" || s.AttemptID == "" || s.CardID == "" || s.ProjectID == "" || s.ConversationID == "" || s.OwnerInstanceID == "" {
+		return fmt.Errorf("task.scope: missing identity")
+	}
+	if s.StreamScope != "conversation:"+s.ConversationID {
+		return fmt.Errorf("task.scope: inconsistent stream")
+	}
+	if s.Revision < 1 || s.Revision > 9007199254740991 || s.SourceSeq < 0 || s.SourceSeq > 9007199254740991 || s.ExecutionEpoch < 0 {
+		return fmt.Errorf("task.revision: unsafe cursor")
+	}
+	switch s.ExecutionState {
+	case "queued", "running", "awaiting_user", "completed", "failed", "interrupted":
+	default:
+		return fmt.Errorf("task.state: unknown")
+	}
+	if (s.ExecutionState == "running" || s.ExecutionState == "awaiting_user") && s.RunID == "" {
+		return fmt.Errorf("task.run: active task requires run")
+	}
+	switch s.OutcomeCompleteness {
+	case "none", "partial", "complete":
+	default:
+		return fmt.Errorf("task.outcome: unknown")
+	}
+	switch s.DeliveryState {
+	case "pending", "available", "reported":
+	default:
+		return fmt.Errorf("task.delivery: unknown")
+	}
+	active := 0
+	for _, f := range s.Findings {
+		if f.ID == "" || f.Revision < 1 {
+			return fmt.Errorf("task.finding: missing identity")
+		}
+		switch f.Kind {
+		case "material_record", "analysis", "provisional", "needs_user", "verified":
+		default:
+			return fmt.Errorf("task.finding.kind: unknown")
+		}
+		switch f.Status {
+		case "active":
+			active++
+		case "corrected", "withdrawn", "superseded":
+		default:
+			return fmt.Errorf("task.finding.status: unknown")
+		}
+	}
+	if active > 12 {
+		return fmt.Errorf("task.findings: too many active")
+	}
+	b, err := json.Marshal(s)
+	if err != nil {
+		return err
+	}
+	if len(b) > 64*1024 {
+		return fmt.Errorf("task.size: exceeds 64KiB")
+	}
+	return nil
+}
+func (s TaskSnapshot) Terminal() bool {
+	return s.ExecutionState == "completed" || s.ExecutionState == "failed" || s.ExecutionState == "interrupted"
+}
+func ValidateWorkCheckpoint(c WorkCheckpointCandidate) error {
+	if strings.TrimSpace(c.CheckpointID) == "" || len(c.CheckpointID) > 128 {
+		return fmt.Errorf("checkpoint.id: required or too long")
+	}
+	switch c.PhaseID {
+	case "reading", "crosscheck", "summarizing", "draft-ready":
+	default:
+		return fmt.Errorf("checkpoint.phase: unknown")
+	}
+	if len(c.Findings) == 0 || len(c.Findings) > 6 {
+		return fmt.Errorf("checkpoint.findings: require 1..6")
+	}
+	keys := map[string]bool{}
+	for _, f := range c.Findings {
+		switch f.Kind {
+		case "material_record", "analysis", "provisional", "needs_user":
+		default:
+			return fmt.Errorf("checkpoint.kind: model cannot grant verified status")
+		}
+		if f.FindingKey == "" || keys[f.FindingKey] || strings.TrimSpace(f.Text) == "" {
+			return fmt.Errorf("checkpoint.finding: empty or duplicate key")
+		}
+		keys[f.FindingKey] = true
+		if len(f.Sources) == 0 {
+			return fmt.Errorf("checkpoint.sources: evidence required")
+		}
+		for _, src := range f.Sources {
+			if src.SourceID == "" || src.BlockIndex < 0 || strings.TrimSpace(src.Quote) == "" {
+				return fmt.Errorf("checkpoint.source: invalid")
+			}
+		}
+		if err := validateWorkFindingEvidence(f.Evidence, len(f.Sources)); err != nil {
+			return err
+		}
+		if (f.Kind == "analysis" || f.Kind == "provisional") && strings.TrimSpace(f.Assumptions) == "" {
+			return fmt.Errorf("checkpoint.assumptions: required")
+		}
+	}
+	b, err := json.Marshal(c)
+	if err != nil {
+		return err
+	}
+	if len(b) > 16*1024 {
+		return fmt.Errorf("checkpoint.size: exceeds 16KiB")
+	}
+	return nil
+}
+func (s TaskSnapshot) SummaryBody() []UINode {
+	labels := map[string]string{"queued": "排队中", "running": "进行中", "awaiting_user": "需要你处理", "completed": "已完成", "failed": "执行失败", "interrupted": "已中断"}
+	var lines []string
+	for _, f := range s.Findings {
+		if f.Status == "active" {
+			line := f.Text
+			if f.Impact != "" {
+				line += "；" + f.Impact
+			}
+			lines = append(lines, line)
+		}
+	}
+	note := strings.Join(lines, "\n")
+	if note == "" {
+		note = TaskSummaryNoFindingsYet
+		if s.ExecutionState == "queued" {
+			note = "尚未开始处理，等待执行资源"
+		}
+		if s.Terminal() {
+			note = s.Conclusion
+			if note == "" {
+				note = "没有可用阶段成果"
+			}
+		}
+	}
+	v := TaskSummaryView{Focus: s.Focus, FindingsNote: note, Remaining: s.Remaining, NextStep: s.NextStep, OutcomeLine: s.Conclusion}
+	var body []UINode
+	if s.Terminal() {
+		body = v.FinishedBody(labels[s.ExecutionState], nil)
+	} else {
+		body = v.RunningBody(labels[s.ExecutionState], nil)
+	}
+	if s.Coverage != "" {
+		body = append(body, UINode{Type: UIKV, KV: [][2]string{{"提供范围", s.Coverage}}})
+	}
+	if len(s.Limitations) > 0 {
+		body = append(body, UINode{Type: UIList, Title: "范围与限制", Items: s.Limitations})
+	}
+	for _, r := range s.ResultRefs {
+		if r.ArtifactID != "" && r.Availability == "available" {
+			body = append(body, UINode{Type: UIDocRef, DocID: r.ArtifactID, Title: r.Name})
+		}
+	}
+	return body
+}
@@ -0,0 +1,147 @@
+package core
+
+import (
+	"encoding/json"
+	"strings"
+	"testing"
+)
+
+// Loss of the optional task object makes refresh discard stage findings.
+func TestTaskSnapshotCardRoundTrip(t *testing.T) {
+	raw := []byte(`{"id":"card1","projectId":"p1","conversationId":"c1","kind":"progress","status":"active","task":{"schemaVersion":1,"taskId":"t1","attemptId":"a1","cardId":"card1","revision":4,"executionState":"running","focus":"核对材料","findings":[{"id":"f1","text":"日期不同","kind":"material_record","status":"active","revision":1}]}}`)
+	var card ActivityCard
+	if err := json.Unmarshal(raw, &card); err != nil {
+		t.Fatal(err)
+	}
+	encoded, err := json.Marshal(card)
+	if err != nil {
+		t.Fatal(err)
+	}
+	var restored map[string]json.RawMessage
+	if err := json.Unmarshal(encoded, &restored); err != nil {
+		t.Fatal(err)
+	}
+	if _, ok := restored["task"]; !ok {
+		t.Fatal("task findings were lost in card persistence")
+	}
+	var task map[string]json.RawMessage
+	if err := json.Unmarshal(restored["task"], &task); err != nil {
+		t.Fatal(err)
+	}
+	if string(task["revision"]) != "4" || string(task["focus"]) != `"核对材料"` {
+		t.Fatal("revision/focus did not survive")
+	}
+}
+
+func validTaskSnapshot() TaskSnapshot {
+	return TaskSnapshot{SchemaVersion: 1, TaskID: "t1", AttemptID: "a1", CardID: "card1", ProjectID: "p1", ConversationID: "c1", OwnerInstanceID: "i1", Revision: 1, ExecutionState: "queued", OutcomeCompleteness: "none", DeliveryState: "pending", StreamScope: "conversation:c1", Title: "整理材料"}
+}
+func TestTaskSnapshotRejectsUnsafeAndUnboundState(t *testing.T) {
+	for _, mutate := range []func(*TaskSnapshot){
+		func(s *TaskSnapshot) { s.TaskID = "" }, func(s *TaskSnapshot) { s.AttemptID = "" },
+		func(s *TaskSnapshot) { s.Revision = 9007199254740992 }, func(s *TaskSnapshot) { s.SourceSeq = -1 },
+		func(s *TaskSnapshot) { s.ExecutionState = "imagined" }, func(s *TaskSnapshot) { s.StreamScope = "conversation:other" },
+		func(s *TaskSnapshot) { s.ExecutionState = "running" },
+	} {
+		s := validTaskSnapshot()
+		mutate(&s)
+		if s.Validate() == nil {
+			t.Errorf("invalid snapshot accepted: %+v", s)
+		}
+	}
+	if err := validTaskSnapshot().Validate(); err != nil {
+		t.Fatal(err)
+	}
+}
+func TestWorkCheckpointRejectsVerifiedAndEmptyEvidence(t *testing.T) {
+	c := WorkCheckpointCandidate{CheckpointID: "cp1", PhaseID: "crosscheck", Findings: []WorkFindingCandidate{{FindingKey: "date", Text: "日期不同", Kind: "verified", Sources: []TaskSourceQuote{{SourceID: "src1", Quote: "六月"}}}}}
+	if err := ValidateWorkCheckpoint(c); err == nil || !strings.Contains(err.Error(), "kind") {
+		t.Fatal("model granted verified status", err)
+	}
+	c.Findings[0].Kind = "material_record"
+	c.Findings[0].Sources = nil
+	if ValidateWorkCheckpoint(c) == nil {
+		t.Fatal("source-free fact accepted")
+	}
+	c.Findings[0].Sources = []TaskSourceQuote{{SourceID: "src1", Quote: "六月"}}
+	if err := ValidateWorkCheckpoint(c); err != nil {
+		t.Fatal(err)
+	}
+	c.Findings[0].Text = strings.Repeat("x", 17000)
+	if ValidateWorkCheckpoint(c) == nil {
+		t.Fatal("unbounded candidate accepted")
+	}
+}
+
+func TestWorkCheckpointEvidenceModesAndBounds(t *testing.T) {
+	base := func() WorkCheckpointCandidate {
+		return WorkCheckpointCandidate{CheckpointID: "cp-evidence", PhaseID: "crosscheck", Findings: []WorkFindingCandidate{{
+			FindingKey: "phase", Text: "待核对", Kind: "needs_user",
+			Sources: []TaskSourceQuote{{SourceID: "total", Quote: "总12项"}, {SourceID: "phase", Quote: "一期8项"}, {SourceID: "dependent", Quote: "其中3项依赖接口"}},
+		}}}
+	}
+	n := func(v int) *int { return &v }
+	total := WorkEvidencePremise{Role: "total", SourceIndex: 0, Value: n(12)}
+	phase := WorkEvidencePremise{Role: "phase", SourceIndex: 1, Value: n(8)}
+	dependent := WorkEvidencePremise{Role: "dependent", SourceIndex: 2, Value: n(3)}
+	for _, tc := range []struct {
+		name     string
+		evidence WorkFindingEvidence
+		valid    bool
+	}{
+		{"direct", WorkFindingEvidence{Mode: "direct_relation", Premises: []WorkEvidencePremise{{Role: "direct_relation", SourceIndex: 0}}}, true},
+		{"difference", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{total, phase}, Result: n(4)}, true},
+		{"unresolved", WorkFindingEvidence{Mode: "unresolved_relation", Premises: []WorkEvidencePremise{total, phase, dependent}}, true},
+		{"unknown mode", WorkFindingEvidence{Mode: "prediction", Premises: []WorkEvidencePremise{total, phase}}, false},
+		{"missing result", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{total, phase}}, false},
+		{"extra result", WorkFindingEvidence{Mode: "unresolved_relation", Premises: []WorkEvidencePremise{total, phase, dependent}, Result: n(3)}, false},
+		{"duplicate role", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{total, total}, Result: n(4)}, false},
+		{"missing premise", WorkFindingEvidence{Mode: "unresolved_relation", Premises: []WorkEvidencePremise{total, phase}}, false},
+		{"negative index", WorkFindingEvidence{Mode: "direct_relation", Premises: []WorkEvidencePremise{{Role: "direct_relation", SourceIndex: -1}}}, false},
+		{"foreign index", WorkFindingEvidence{Mode: "direct_relation", Premises: []WorkEvidencePremise{{Role: "direct_relation", SourceIndex: 3}}}, false},
+		{"negative value", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{{Role: "total", SourceIndex: 0, Value: n(-1)}, phase}, Result: n(4)}, false},
+		{"large value", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{{Role: "total", SourceIndex: 0, Value: n(1000001)}, phase}, Result: n(4)}, false},
+		{"direct value", WorkFindingEvidence{Mode: "direct_relation", Premises: []WorkEvidencePremise{{Role: "direct_relation", SourceIndex: 0, Value: n(1)}}}, false},
+		{"missing value", WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{{Role: "total", SourceIndex: 0}, phase}, Result: n(4)}, false},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			c := base()
+			c.Findings[0].Evidence = &tc.evidence
+			err := ValidateWorkCheckpoint(c)
+			if (err == nil) != tc.valid {
+				t.Fatalf("valid=%v err=%v", tc.valid, err)
+			}
+		})
+	}
+	c := base()
+	c.Findings[0].Evidence = &WorkFindingEvidence{Mode: "difference", Premises: []WorkEvidencePremise{total, phase}, Result: n(4)}
+	c.Findings[0].Text = strings.Repeat("字", 5500)
+	if err := ValidateWorkCheckpoint(c); err == nil || !strings.Contains(err.Error(), "16KiB") {
+		t.Fatal("evidence bypassed checkpoint size cap", err)
+	}
+}
+func TestTaskSnapshotBodyPreservesLimitsAndHonestStates(t *testing.T) {
+	s := validTaskSnapshot()
+	s.Remaining = "图片内容未纳入"
+	s.ExecutionState = "failed"
+	s.Conclusion = "处理失败，已有发现可查看"
+	s.Findings = []TaskFinding{{ID: "f1", Text: "金额缺失", Kind: "material_record", Status: "active", Revision: 1}}
+	raw, _ := json.Marshal(s.SummaryBody())
+	for _, want := range []string{"执行失败", "图片内容未纳入", "金额缺失", "处理失败"} {
+		if !strings.Contains(string(raw), want) {
+			t.Errorf("lost critical summary %s", want)
+		}
+	}
+}
+
+func TestTaskSnapshotUnstartedAndFinishedDoNotClaimReading(t *testing.T) {
+	for _, state := range []string{"queued", "completed", "interrupted"} {
+		s := validTaskSnapshot()
+		s.ExecutionState = state
+		s.Conclusion = "处理已结束"
+		b, _ := json.Marshal(s.SummaryBody())
+		if strings.Contains(string(b), "正在读取") {
+			t.Errorf("%s falsely claims active reading", state)
+		}
+	}
+}
@@ -149,6 +149,7 @@ export interface ActivityCard {
  id: ID; projectId: ID; conversationId?: ID;
  sourceRunId?: ID; anchorMessageId?: ID;
  kind: CardKind; title: string; status: CardStatus; tone?: string;
+  task?: TaskSnapshot;
  body?: UINode[]; actions?: { id: string; label: string; style?: 'primary' | 'default' | 'danger'; command?: string }[];
  pinned?: boolean; createdAt: string; updatedAt: string; work?: WorkCard;
 }
@@ -350,7 +351,7 @@ export interface PluginPageDecl {

 export interface ParsedDocument {
  meta: {
-    filename: string; mime?: string; parser: string; confidence?: number;
+    filename: string; mime?: string; parser: string; confidence?: number; origin?: DocumentOrigin;
    crossCheck?: { with: string; diffBlocks: number; verdict: 'agree' | 'differ' };
    parsedAt: string;
  };
@@ -413,3 +414,28 @@ export interface SurfaceMeta {
  conversationId?: ID;
  updatedAt: string;
 }
+
+// Persistent work projection; host-owned identities and revisions.
+export interface DocumentOrigin { artifactId: ID; contentHash: string; parserRevision: string }
+export interface TaskPhase { id: string; label: string; observedAt?: string }
+export interface TaskSourceRef { id: string; artifactId: ID; contentHash: string; parserRevision: string; blockIndex: number; label: string; quote?: string; availability?: string }
+export interface TaskResultRef { artifactId?: ID; surfaceId?: ID; version?: string; contentHash?: string; availability: string; name?: string }
+export interface TaskFinding { id: string; text: string; impact?: string; kind: string; status: string; revision: number; sources: TaskSourceRef[]; assumptions?: string; corrects?: string }
+export interface TaskSnapshot {
+	inputs?: TaskSourceRef[]
+ limitations?: string[]
+ schemaVersion: number; taskId: ID; attemptId: ID; cardId: ID; projectId: ID; conversationId: ID;
+ sourceMessageId?: ID; parentRunId?: ID; runId?: ID; ownerInstanceId: string; executionEpoch?: number;
+ revision: number; sourceSeq: number; streamScope: string; inputRevision?: string; title: string; goal?: string;
+ executionState: 'queued'|'running'|'awaiting_user'|'completed'|'failed'|'interrupted'; outcomeCompleteness: 'none'|'partial'|'complete'; deliveryState: 'pending'|'available'|'reported'; phase: TaskPhase;
+ focus?: string; coverage?: string; remaining?: string; nextStep?: string; conclusion?: string;
+ findings?: TaskFinding[]; actions?: UIAction[]; resultRefs?: TaskResultRef[]; lastActivityAt?: string; lastMeaningfulUpdateAt?: string;
+ checkpointKeys?: Record<string,string>; checkpointCount?: number; checkpointSeq?: number;
+}
+export interface TaskSourceQuote { sourceId: string; blockIndex: number; quote: string }
+interface WorkEvidencePremise { role: "total" | "phase" | "dependent" | "direct_relation"; sourceIndex: number; value?: number }
+interface WorkFindingEvidence { mode: "direct_relation" | "difference" | "unresolved_relation"; premises: WorkEvidencePremise[]; result?: number }
+export interface WorkFindingCandidate { findingKey: string; text: string; impact?: string; kind: string; assumptions?: string; corrects?: string; sources: TaskSourceQuote[]; evidence?: WorkFindingEvidence }
+export interface WorkCheckpointCandidate { checkpointId: string; phaseId: string; focus?: string; remaining?: string; nextStep?: string; findings: WorkFindingCandidate[] }
+
+export interface UIAction { id: string; label: string; style?: string; command?: string }
Author	SHA1	Message	Date
weichen	6fe40b0342	Merge pull request 'Add bounded evidence to work checkpoint candidates' (#31 ) from feature/fast-slow-u2-u3 into main governance / governance (push) Successful in 16s Details	2026-10-19 09:20:24 +00:00
weichen	0cf03446c4	Merge the fast-slow U2/U3 branch history after syncing with main governance / governance (pull_request) Successful in 16s Details	2026-10-19 02:59:18 +00:00
agent-exec-transport	cbd3d3e3d0	agent-exec transport 68889ca8d4b0eec4cfe89d0cc80b6516/5ca3b7798211	2026-10-19 02:59:03 +00:00
weichen	5cae3a6e30	Merge pull request 'Adopt the shared governance kit (development-governance-v2)' (#35 ) from feature/governance-adoption into main governance / governance (push) Successful in 17s Details	2026-10-17 08:31:36 +00:00
weichen	b240077300	Close and archive the governance adoption card governance / governance (pull_request) Successful in 16s Details Why: the adoption PR is reviewed and its candidate HEAD passed CI; the card closure is folded into the PR so no separate close PR is needed. Evidence: CI GATE on 18f617d4e8cd63d172a5ab9519271831beb546db; independent Review REPLAY (cursor-agent-stable --mode ask 全新只读会话（fresh_session=true，session id not-observable；codex/pi 因 Token Plan 429 不可用）VERDICT: APPROVE，无阻塞项). Impact: task card only (moved to .task/archive/). Refs: https://git-home.u1680832.nyat.app:30319/weichen/Agent-System-Core/pulls/35	2026-10-17 08:25:12 +00:00
weichen	18f617d4e8	Record the adoption unit run and positive/negative evidence governance / governance (pull_request) Successful in 23s Details Why: Phase 6 requires one positive executor run and the negative cases (missing config, bad receipt, out-of-scope path, unknown side effect) before adoption becomes active. Evidence: .task/evidence/governance-adoption/ (SELF): governance check, agent-exec validate and adoption cases; this commit passed the generated pre-commit hook (admit). Impact: task card and evidence logs only. Refs: .task/feature-governance-adoption.md	2026-10-17 07:15:37 +00:00
agent-exec-transport	1ada9955d5	agent-exec transport e64b5afee9fabb0925523057eb540b68/8399a44af60d	2026-10-17 06:55:22 +00:00
weichenandCursor	a160b03c7b	Record the adoption unit run and positive/negative evidence Why: Phase 6 requires one positive executor run and the negative cases (missing config, bad receipt, out-of-scope path, unknown side effect) before adoption becomes active. Evidence: .task/evidence/governance-adoption/ (SELF): governance check all ok; agent-exec validate ok; adoption cases ALL PASS. The generated pre-commit hook refused this commit because scripts/dm resolves the DM checkout through the hook's GIT_DIR; fixed in Development-Manager before this repository re-pins. Impact: task card and evidence logs only. Refs: .task/feature-governance-adoption.md Co-authored-by: Cursor <cursoragent@cursor.com>	2026-10-17 06:19:48 +00:00
agent-exec-transport	98cfc6408d	agent-exec transport 9a7ec13541404ada96c0f8bfc8d77058/a09c2ed0bb4b	2026-10-17 05:55:46 +00:00
weichen	dde4f359a8	Bootstrap AGENT-EXECUTION.md and the governance adoption card Why: development-governance-v2 Phase 6 adopts the shared governance kit in every formal repository; the executor needs a valid AGENT-EXECUTION.md before any implementation unit can run. Evidence: generated by dm-server governance init at Development-Manager 42e675fec8e0069646319fcd298e61d03d8cbfd5; dm-server agent-exec validate passes. Impact: configuration and task card only; no code or CI change yet. Refs: Agent-System-Dev docs/superpowers/plans/2026-10-16-development-governance-v2-master.md	2026-10-17 05:51:28 +00:00
weichen	35560aa5a3	Merge pull request 'Keep complete background results and optional decisions on work cards' (#34 ) from feature/workbench-flow-completion into main	2026-10-15 08:18:43 +00:00
weichen	b4743949e5	Keep U3 TypeScript evidence helpers local Why: Generated mirrors gained unused named exports and tripped the frontend knip ratchet. Evidence: Exact Dev combination failed at 53 unused exports over 52 baseline; Core Go tests passed after narrowing the TS export surface. Impact: Preserve WorkFindingCandidate.evidence shape while avoiding unused exported helper names. Refs: Dev task feature/fast-slow-u2-u3-plan; failed combo receipt 275bebe2	2026-10-13 16:48:01 +00:00
weichen	c722c16626	Validate optional evidence in work checkpoint candidates Why: U3 needs bounded, source-indexed premises before the host can judge numeric and membership claims. Evidence: TestWorkCheckpointEvidenceModesAndBounds failed on missing types, then go test ./... -count=1 and git diff --check passed. Impact: Extend only the candidate wire contract; persisted task snapshots and display kinds stay unchanged. Refs: Dev task feature/fast-slow-u2-u3-plan; r0.3 evidence implementation plan Task 1	2026-10-13 16:03:33 +00:00
weichen	c55ebbcb96	Merge remote-tracking branch 'origin/main' into feature/fast-slow-u2-u3	2026-10-13 10:19:06 +00:00
weichen	beafd7537a	Merge scoped settings contracts from current main Why: Preserve upstream settings authority and current component admission. Evidence: Exact authoritative main object fetched before merge; composition replay follows. Impact: Retain Task contracts alongside scoped settings. Refs: Agent-System-Dev U2/U3 delivery	2026-10-13 05:28:02 +00:00
weichen	6e385743e4	Retain coverage limits in legacy task summaries Why: Older clients must see the same coverage limits as structured task cards. Evidence: Core suite and composition checks follow this projection update. Impact: Projects optional limitations into existing list nodes. Refs: fast-slow-u2-u3 Task 6	2026-10-13 02:08:00 +00:00
weichen	dbe1f544d7	Keep material coverage limits separate from outcome claims Why: A completed run can still lack readable or fully provided source material. Evidence: Composition validation follows the backend and UI consumers. Impact: Adds optional host-owned limitations without changing candidate authority. Refs: fast-slow-u2-u3 Tasks 4 through 6	2026-10-13 01:59:40 +00:00
weichen	124309b985	Retain trusted input bindings for task result reuse Why: Completed tasks require current source authorization and content validation. Evidence: Optional fields preserve existing contracts; composition verification follows. Impact: Adds host-owned input source references with no new model authority. Refs: fast-slow-u2-u3 Tasks 2 and 4	2026-10-13 01:35:53 +00:00
weichen	06f2723161	Add persistent work snapshots and checkpoint contracts Why: preserve scoped work findings and truthful activity across reloads. Evidence: task JSON loss and validation/state regressions RED to GREEN; complete Core suite. Impact: optional card/document wire fields and bounded candidate validation; old Body remains supported. Refs: fast-slow U2/U3 Task1, R03 R05 R09 R17	2026-10-13 01:04:56 +00:00