@fitlab-ai/agent-infra 0.9.13 → 0.9.15

package/templates/.agents/skills/complete-manual-validation/SKILL.en.md

@@ -9,14 +9,18 @@ description: >
 # Complete Manual Validation
 > `--agent` values are defined in `.agents/rules/task-management.md` under “Collaborator Token Specification”.
 
-Lifecycle events require explicit trigger data: use `{trigger-initiator}=orchestrator` for orchestration and `model` otherwise; `{request-id}` is a stable single-line identifier for this task and artifact round, and `{reason-code}` is `user-request` or `validation-rerun`. Reuse the same values for started and completed.
+The transaction coordinator appends lifecycle events and the final PR summary state. The maintainer's validation summary and PR manual-validation comment are the human-validation authority; the transaction receipt provides state-transition and recovery consistency.
 
 
 ## Boundary / Critical Rules
 
+### Persisted Report Evidence
+
+Before generating the validation completion report, read `.agents/rules/evidence-reporting.md`. Record commands, scope, structured results, actual conclusions, and uncovered parts for state checks and synchronization; keep the stricter basename-only and sanitized-result boundaries for manual validation.
+
 - This skill closes the manual-validation status in an existing PR summary comment; it does not create a parallel ordinary validation comment.
 - It must write `manual-validation.md` or `manual-validation-r{N}.md` so later PR summary refreshes can reuse the validation result.
-- If the `sync-pr` summary comment is missing, fail instead of creating a partial fallback summary.
+- If the `sync-pr` summary comment is missing, fail instead of creating a partial fallback summary; before the receipt, completion log, and final summary are all committed, the remote summary may only be pending/non-pass.
 - Before generating manual-validation artifact Markdown that will be synced to an Issue, read `.agents/rules/sync-content-generation.md` and follow its generator-side constraints; Issue sync remains transparent and does not parse or rewrite the body.
 - After this skill runs, update `task.md` immediately.
 
@@ -26,7 +30,7 @@ Version stamp rule: when creating or updating `task.md` frontmatter, read `.agen
 
 After loading workflow / skill / rules instructions, and before any task-state judgment or user-visible conclusion, run the state check first. Reading instruction files does not count as an external-state action or conclusion.
 
-Run these commands and paste the raw output into this round's `## State Check` section:
+Run these commands and record the task/artifact scope, key result, and uncovered parts in this round's `## State Check` section; do not paste complete directory listings or task tails on normal success. Retain decisive raw lines only for failures, blocking conditions, identity mismatches, or disputes:
 
 ```bash
 agent-infra-internal task-snapshot {task-id} --format text
@@ -38,10 +42,6 @@ agent-infra-internal task-snapshot {task-id} --format text
 
 > Resolve the task reference, then confirm that the task is in a state or directory supported by this skill and that `task.md` exists; if it cannot be located, handle it as a missing task and stop.
 
-## Step Start: Write the started Marker
-
-After resolving the artifact context and before this round's first artifact action, run `agent-infra-internal task-event {task-id} manual-validation.started --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code}` and record the returned `artifactContext`.
-
 ## Steps
 
 ### 1. Parse Arguments
@@ -66,33 +66,49 @@ Stop if the task is missing, the validation summary is missing, or no valid PR c
 
 ### 3. Resolve the Artifact Context
 
-Run `agent-infra-internal task-artifact {task-id} inspect --family manual-validation`. Continue only for `ready`; take the round and `{manual-validation-artifact}` from `next.round` / `next.name`. Do not scan rounds or construct names in the skill. Then run the started event and verify the returned identity.
+Run `agent-infra-internal task-artifact {task-id} inspect --family manual-validation`. Continue only for `ready`; take the round and `{manual-validation-artifact}` from `next.round` / `next.name`. Do not scan rounds or construct names in the skill. The transaction coordinator owns the started event.
+
+### 4. Register Manual Validation Start
+
+Before creating any manual-validation artifact, obtain the canonical summary body from `platform-pr summary-context`, then call the transaction coordinator in prepare mode:
+
+```bash
+agent-infra-internal manual-validation transaction {task-id} --prepare \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --agent {standard-agent-token}
+```
 
-### 4. Update the PR Summary
+Prepare must idempotently record `manual-validation.started` before writing the prepared transaction; on failure, stop without creating the artifact.
+
+### 5. Create the Manual Validation Artifact
+
+Before this step, read `reference/report-template.md`. After `started` is recorded, create `{manual-validation-artifact}` and record the state check, validation verdict, validation scope, validation details, and expected PR summary synchronization result; the transaction coordinator validates and commits this artifact in the next step.
+
+### 6. Update the PR Summary
 
 Before this step, read:
 - `.agents/rules/issue-sync.md`
 - `.agents/rules/pr-sync.md`
 - `reference/summary-update.md`
 
-Follow `reference/summary-update.md` to validate the PR binding, obtain canonical inputs from `platform-pr summary-context`, and update the manual-validation section through `platform-pr summary-sync` to `### ✅ Manual Validation Passed`.
+Follow `reference/summary-update.md` to validate the PR binding, obtain canonical inputs from `platform-pr summary-context`, and call the transaction coordinator once:
 
-### 5. Create the Manual Validation Artifact
+```bash
+agent-infra-internal manual-validation transaction {task-id} \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json \
+  --agent {standard-agent-token} --result no_op
+```
 
-Before this step, read `reference/report-template.md`. Create `{manual-validation-artifact}` and record:
-- State check
-- Validation verdict
-- Validation scope
-- Validation details
-- PR summary sync result
+The coordinator owns pending summary, receipt, completion log, final promotion, and post-write verification; it internally uses the controlled `agent-infra-internal task-event {task-id}` route. Do not call final `summary-sync` or `manual-validation.completed` separately. Raw evidence files from other hosts are not required.
 
-### 6. Update task.md
+### 7. Update task.md
 
-Run `agent-infra-internal task-event {task-id} manual-validation.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {manual-validation-artifact} --summary-result "{summary-result}"`. The core keeps `current_step` unchanged while atomically recording the implementation-notes link, metadata, and done log.
+After the transaction coordinator succeeds, the core has atomically recorded `manual-validation.completed` with the same transaction/receipt/head identity. Do not append the Activity Log manually.
 
 If the task has a valid `issue_number`, run `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}`, then `agent-infra-internal platform-comment sync {task-id} --kind artifact --artifact {manual-validation-artifact} --agent {standard-agent-token}`.
 
-### 7. Verification Gate
+### 8. Verification Gate
 
 Run:
 
@@ -105,7 +121,7 @@ Handle the result:
 - Exit code 1 -> fix the reported problem and rerun
 - Exit code 2 -> stop and report that manual intervention is required
 
-### 8. Tell the User
+### 9. Tell the User
 
 Report:
 - Artifact path

package/templates/.agents/skills/complete-manual-validation/SKILL.zh-CN.md

@@ -9,14 +9,18 @@ description: >
 # 完成人工验证
 > `--agent` 取值见 `.agents/rules/task-management.md`「合作者 token 规范」。
 
-生命周期事件必须携带显式触发信息：编排调用使用 `{trigger-initiator}=orchestrator`，否则使用 `model`；`{request-id}` 是本任务与本轮产物的稳定单行标识，`{reason-code}` 使用 `user-request` 或 `validation-rerun`；started 与 completed 使用同一组值。
+事务协调器统一追加 lifecycle 事件和 PR 摘要最终状态。维护者提供的验证说明和 PR 摘要留言就是人工验证依据；事务回执负责保证状态转换和恢复一致性。
 
 
 ## 行为边界 / 关键规则
 
+### 持久化报告证据
+
+生成验证完成报告时，先读取 `.agents/rules/evidence-reporting.md`。状态核对和同步结果记录命令、范围、结构化结果、实际结论和未覆盖部分；继续遵守人工验证的 basename-only 与 sanitized result 脱敏边界。
+
 - 本技能用于收尾已有 PR 摘要评论中的人工校验状态，不创建并行的普通验证留言。
 - 必须写入 `manual-validation.md` 或 `manual-validation-r{N}.md`，让后续 PR 摘要刷新可复用人工验证结果。
-- 找不到 `sync-pr` 摘要评论时失败，不创建部分摘要兜底。
+- 找不到 `sync-pr` 摘要评论时失败，不创建部分摘要兜底；receipt、通过日志和 final summary 未全部提交前，远端只允许 pending/non-pass 状态。
 - 生成会同步到 Issue 的人工验证 artifact Markdown 前，先读取 `.agents/rules/sync-content-generation.md` 并遵循其中的生成端约束；Issue 同步保持透明，不解析或改写正文。
 - 执行本技能后必须立即更新 `task.md`。
 
@@ -26,7 +30,7 @@ description: >
 
 在加载 workflow / skill / rules 指令之后、做任何任务状态判断或用户可见结论之前，必须先执行状态核对。指令类文件读取不算对外动作或结论。
 
-运行以下命令，并把原文粘贴到本轮产物的 `## 状态核对` 段：
+运行以下命令，并在本轮产物的 `## 状态核对` 段记录任务/产物范围、关键结果和未覆盖部分；正常成功不粘贴完整目录清单或 `task.md` 尾部。失败、阻塞、身份不一致或争议时，附决定性原文行：
 
 ```bash
 agent-infra-internal task-snapshot {task-id} --format text
@@ -38,10 +42,6 @@ agent-infra-internal task-snapshot {task-id} --format text
 
 > 解析任务引用，并确认任务位于本技能支持的状态或目录且存在 `task.md`；无法定位时按未找到任务处理并停止。
 
-## 步骤开始：声明 started 事件
-
-确认前置条件和产物上下文后、本轮第一个产出动作之前执行 `agent-infra-internal task-event {task-id} manual-validation.started --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code}`，并以返回的 `artifactContext` 记录本轮身份。
-
 ## 执行步骤
 
 ### 1. 解析入参
@@ -66,33 +66,49 @@ complete-manual-validation [--task <ref> | -t <ref>] [{pr-ref}] {verification-su
 
 ### 3. 解析产物上下文
 
-运行 `agent-infra-internal task-artifact {task-id} inspect --family manual-validation`。仅当结果为 `ready` 时继续；从 `next.round` / `next.name` 取得本轮 round 与 `{manual-validation-artifact}`。不得自行扫描轮次或拼装文件名。随后执行 started 事件并复核返回身份。
+运行 `agent-infra-internal task-artifact {task-id} inspect --family manual-validation`。仅当结果为 `ready` 时继续；从 `next.round` / `next.name` 取得本轮 round 与 `{manual-validation-artifact}`。不得自行扫描轮次或拼装文件名。事务协调器负责 started 事件。
+
+### 4. 登记人工验证开始
+
+在创建任何人工验证 artifact 前，先从 `platform-pr summary-context` 取得 canonical 摘要正文，然后调用事务协调器的 prepare 模式：
+
+```bash
+agent-infra-internal manual-validation transaction {task-id} --prepare \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --agent {standard-agent-token}
+```
 
-### 4. 更新 PR 摘要
+prepare 必须先幂等登记 `manual-validation.started`，再写入 prepared transaction；失败时停止，不创建 artifact。
+
+### 5. 创建人工验证产物
+
+执行此步骤前，先读取 `reference/report-template.md`。在 started 已登记后创建 `{manual-validation-artifact}`，记录状态核对、验证结论、验证范围、验证详情和预期 PR 摘要同步结果；事务协调器将在后续步骤校验并提交这份 artifact。
+
+### 6. 更新 PR 摘要
 
 执行此步骤前，先读取：
 - `.agents/rules/issue-sync.md`
 - `.agents/rules/pr-sync.md`
 - `reference/summary-update.md`
 
-按 `reference/summary-update.md` 校验 PR 绑定，从 `platform-pr summary-context` 取得 canonical 输入，并通过 `platform-pr summary-sync` 把人工校验段更新为 `### ✅ 人工验证已通过`。
+按 `reference/summary-update.md` 校验 PR 绑定，从 `platform-pr summary-context` 取得 canonical 输入，并调用一次 transaction coordinator：
 
-### 5. 创建人工验证产物
+```bash
+agent-infra-internal manual-validation transaction {task-id} \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json \
+  --agent {standard-agent-token} --result no_op
+```
 
-执行此步骤前，先读取 `reference/report-template.md`。创建 `{manual-validation-artifact}`，记录：
-- 状态核对
-- 验证结论
-- 验证范围
-- 验证详情
-- PR 摘要同步结果
+coordinator 负责 pending summary、receipt、通过日志、final promotion 和 post-write verification；内部受控调用 `agent-infra-internal task-event {task-id}`，不要分别调用 final `summary-sync` 或 `manual-validation.completed`。不要求调用方提供其他主机上的原始 evidence 文件。
 
-### 6. 更新 task.md
+### 7. 更新 task.md
 
-执行 `agent-infra-internal task-event {task-id} manual-validation.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {manual-validation-artifact} --summary-result "{summary-result}"`，由核心在保持 `current_step` 不变的同时原子登记实现备注链接、时间/版本和完成日志。
+transaction coordinator 成功后，核心已使用同一 transaction/receipt/head identity 原子登记 `manual-validation.completed`；不要手工补写 Activity Log。
 
 如任务存在有效 `issue_number`，调用 `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}`，再调用 `agent-infra-internal platform-comment sync {task-id} --kind artifact --artifact {manual-validation-artifact} --agent {standard-agent-token}`。
 
-### 7. 完成校验
+### 8. 完成校验
 
 运行完成校验：
 
@@ -105,7 +121,7 @@ agent-infra-internal task-verify {task-id} manual-validation.completed --artifac
 - 退出码 1 -> 修复问题后重新运行
 - 退出码 2 -> 停止并告知需要人工介入
 
-### 8. 告知用户
+### 9. 告知用户
 
 输出：
 - 产物路径

package/templates/.agents/skills/complete-manual-validation/config/verify.en.json

@@ -37,7 +37,7 @@
       "when": "pr_fact_bound",
       "expected_pr_comment_marker_key": "prSummary",
       "expected_pr_comment_required_patterns": [
-        "^### ✅ Manual Validation Passed$"
+        "^### ✅ (?:人工验证已通过|Manual Validation Passed)$"
       ],
       "verify_task_comment_content": true,
       "verify_issue_type": true,

package/templates/.agents/skills/complete-manual-validation/config/verify.zh-CN.json

@@ -37,7 +37,7 @@
       "when": "pr_fact_bound",
       "expected_pr_comment_marker_key": "prSummary",
       "expected_pr_comment_required_patterns": [
-        "^### ✅ 人工验证已通过$"
+        "^### ✅ (?:人工验证已通过|Manual Validation Passed)$"
       ],
       "verify_task_comment_content": true,
       "verify_issue_type": true,

package/templates/.agents/skills/complete-manual-validation/reference/report-template.en.md

@@ -7,6 +7,7 @@ Read this file before creating `manual-validation.md` / `manual-validation-r{N}.
 
 - **Validation Round**: Round {N}
 - **Artifact File**: `manual-validation.md`
+- **Human Validation Basis**: Maintainer's PR validation comment and validation summary
 
 ## State Check
 
@@ -15,6 +16,7 @@ $ git status -s
 $ ls -la .agents/workspace/active/{task-id}/
 $ tail .agents/workspace/active/{task-id}/task.md
 ```
+> Follow `.agents/rules/evidence-reporting.md`: record task scope, key result, and uncovered parts; do not paste the complete directory listing or task tail for normal success.
 
 ## Validation Verdict
 
@@ -38,6 +40,7 @@ $ tail .agents/workspace/active/{task-id}/task.md
 - Result: {summary-result}
 - Summary comment: {comment-id or URL}
 - Updated status: `### ✅ Manual Validation Passed`
+- Transaction: `{transaction-id}`; receipt=`{receipt-digest}`; head=`{pr-head-sha}`
 ````
 
 ## Filling Rules
@@ -46,3 +49,4 @@ $ tail .agents/workspace/active/{task-id}/task.md
 - Derive `Validation Scope` from the PR summary's manual-verification section.
 - `PR Summary Sync` must record success, skip, or failure.
 - Only write a passing artifact after the summary comment is updated or skipped because there is no diff.
+- Successful synchronization records the command, scope, structured result, and actual conclusion; failures, blocking conditions, or disputes retain decisive excerpts and manual-validation redaction boundaries.

package/templates/.agents/skills/complete-manual-validation/reference/report-template.zh-CN.md

@@ -7,6 +7,7 @@
 
 - **验证轮次**：Round {N}
 - **产物文件**：`manual-validation.md`
+- **人工验证依据**：维护者提供的 PR 验证留言和验证说明
 
 ## 状态核对
 
@@ -15,6 +16,7 @@ $ git status -s
 $ ls -la .agents/workspace/active/{task-id}/
 $ tail .agents/workspace/active/{task-id}/task.md
 ```
+> 按 `.agents/rules/evidence-reporting.md` 记录任务范围、关键结果和未覆盖部分；正常成功不粘贴完整目录清单或 task tail。
 
 ## 验证结论
 
@@ -38,6 +40,7 @@ $ tail .agents/workspace/active/{task-id}/task.md
 - 结果：{summary-result}
 - 摘要评论：{comment-id 或 URL}
 - 更新状态：`### ✅ 人工验证已通过`
+- Transaction：`{transaction-id}`；receipt=`{receipt-digest}`；head=`{pr-head-sha}`
 ````
 
 ## 填写规则
@@ -46,3 +49,4 @@ $ tail .agents/workspace/active/{task-id}/task.md
 - `验证范围` 来自 PR 摘要评论中的 `### ⚠️ 需人工校验` 段。
 - `PR 摘要同步` 必须记录成功、跳过或失败结果。
 - 只有在摘要评论已成功更新或无 diff 跳过时，才写入通过产物。
+- 成功同步记录命令、范围、结构化结果和实际结论；失败、阻塞或争议保留决定性摘录，并继续遵守人工验证脱敏边界。

package/templates/.agents/skills/complete-manual-validation/reference/summary-update.en.md

@@ -4,14 +4,23 @@ Run only for a valid `{task-id}` whose task.md binds a verified `pr_delivery_fac
 
 Follow `.agents/rules/pr-sync.md` for the shared summary structure and failure semantics.
 
-1. After the canonical `manual-validation*` artifact is written, run `agent-infra-internal platform-pr summary-context {task-id}`.
-2. Rebuild the summary from canonical inputs and render `### ✅ Manual Validation Passed` with the validation time and explanation.
-3. Write the plain body with exactly one `<!-- canonical-pr-change-report -->` placeholder to a temporary file and run:
+1. Run `agent-infra-internal platform-pr summary-context {task-id}` to obtain the canonical summary body; the maintainer's PR validation comment and validation summary are the human-validation authority.
+2. Before creating the canonical `manual-validation*` artifact, call the transaction coordinator in prepare mode to record `started` before writing the prepared transaction:
 
 ```bash
-agent-infra-internal platform-pr summary-sync {task-id} \
-  --agent {standard-agent-token} --body-file {summary-body-file} \
-  --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json --result no_op
+agent-infra-internal manual-validation transaction {task-id} --prepare \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --agent {standard-agent-token}
 ```
 
-The core owns the rendered report, marker/authoritative PR-head wrapping, pagination, and in-place reconciliation. If context says no manual validation is required, stop without marking it passed.
+3. After prepare succeeds, write the canonical `manual-validation*` artifact, write the pending plain body with exactly one `<!-- canonical-pr-change-report -->` placeholder to a temporary file, and run the transaction coordinator:
+
+```bash
+agent-infra-internal manual-validation transaction {task-id} \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} \
+  --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json \
+  --agent {standard-agent-token} --result no_op
+```
+
+The core owns the rendered report, marker/authoritative PR-head wrapping, pagination, pending-first ordering, receipt, completion log, final promotion, and in-place reconciliation. Before the receipt and completion log exist, the summary must not contain `### ✅ Manual Validation Passed`. If context says no manual validation is required, stop without marking it passed.

package/templates/.agents/skills/complete-manual-validation/reference/summary-update.zh-CN.md

@@ -4,16 +4,25 @@
 
 摘要结构与失败语义统一遵循 `.agents/rules/pr-sync.md`。
 
-1. 在 canonical `manual-validation*` artifact 落盘后，调用 `agent-infra-internal platform-pr summary-context {task-id}`。
-2. 从返回的最新产物重新聚合摘要，将人工校验段渲染为 `### ✅ 人工验证已通过`，写明验证时间和说明。
-3. 把只含一次 `<!-- canonical-pr-change-report -->` 的纯正文写入临时文件，调用：
+1. 调用 `agent-infra-internal platform-pr summary-context {task-id}`，取得 canonical 摘要正文；维护者提供的 PR 人工验证留言和验证说明作为完成人工校验依据。
+2. 在创建 canonical `manual-validation*` artifact 前，调用 transaction coordinator 的 prepare 模式，先登记 started，再写 prepared transaction：
 
 ```bash
-agent-infra-internal platform-pr summary-sync {task-id} \
-  --agent {standard-agent-token} --body-file {summary-body-file} \
-  --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json --result no_op
+agent-infra-internal manual-validation transaction {task-id} --prepare \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} --agent {standard-agent-token}
 ```
 
-marker、报告段、权威 PR head、分页查找和原地更新由 core 负责。若当前 context 表明无需人工校验，则停止并返回 `summary failed: no manual validation required`，不误标通过。
+3. prepare 成功后写入 canonical `manual-validation*` artifact，并把只含一次 `<!-- canonical-pr-change-report -->` 的 pending 正文写入临时文件，调用 transaction coordinator：
+
+```bash
+agent-infra-internal manual-validation transaction {task-id} \
+  --artifact {manual-validation-artifact} \
+  --summary-file {summary-body-file} \
+  --change-report-file .agents/workspace/active/{task-id}/pr-change-report.json \
+  --agent {standard-agent-token} --result no_op
+```
+
+marker、报告段、权威 PR head、分页查找、pending-first 顺序、receipt、通过日志和 final promotion 由 core 负责。若当前 context 表明无需人工校验，则停止并返回 `summary failed: no manual validation required`，不误标通过。receipt/log 前不得出现 `### ✅ 人工验证已通过`。
 
 结果回传：`summary updated`、`summary skipped (no diff)` 或 `summary failed: <reason>`。

package/templates/.agents/skills/complete-task/SKILL.en.md

@@ -14,6 +14,10 @@ Host finalization uses receipt v2 with an immutable `receiptId`, monotonic `revi
 
 ## Boundary / Critical Rules
 
+### Persisted Report Evidence
+
+Before generating a completion report or synchronized content, read `.agents/rules/evidence-reporting.md`. Successful checks record the command, scope, status or structured result, actual result, and uncovered parts; failures, blocking conditions, or disputes retain a reproducible entry point, exact location, and decisive excerpt.
+
 - This command updates task metadata AND physically moves the task directory
 - Do not move a task that has incomplete workflow steps unless forced
 - The entry point accepts optional `--external-pr <N>` only to select among ambiguous external-delivery candidates; it never bypasses identity or platform gates
@@ -24,7 +28,7 @@ Version stamp rule: when creating or updating `task.md` frontmatter, read `.agen
 
 After loading workflow / skill / rules instructions, and before any task-state judgment or user-visible conclusion, run the state check first. Reading instruction files does not count as an external-state action or conclusion.
 
-Run these commands and paste the raw output into this round's `## State Check` section:
+Run these commands and record the task/artifact scope, key result, and uncovered parts in this round's `## State Check` section; do not paste complete directory listings or task tails on normal success. Retain decisive raw lines only for failures, blocking conditions, identity mismatches, or disputes:
 
 ```bash
 agent-infra-internal task-snapshot {task-id} --format text
@@ -100,7 +104,7 @@ Run /create-pr --task {task-ref} first, then complete; --skip-pr is not accepted
 Before lifecycle, verify only the hard gates:
 - [ ] Task identity, active state, concurrency locks, and local atomic lifecycle operations are available
 - [ ] Required-PR delivery is satisfied when `prFlow=required`
-- [ ] Other business evidence (workflow, review, commit, tests, disagreement ledger, manual validation, and platform sync) is recorded or can be checked after lifecycle
+- [ ] Other business evidence (workflow, review, commit, tests, disagreement ledger, manual validation, and platform sync) is recorded or can be checked after lifecycle; manual validation requires a committed transaction, receipt, current artifact digest, and an identity-bearing completion log
 
 > **⚠️ Prerequisite Branch Check — you must decide whether to continue or stop before proceeding:**
 >
@@ -123,7 +127,7 @@ If a hard gate is not met, stop immediately and do not execute Steps 3-8.
 ### 3. Complete Business-Only Content
 
 Update only content that the lifecycle core does not own:
-- Add or update the `## State Check` section with the raw Step 0 audit command output, including `$ ` prompt lines, before `## Activity Log`
+- Add or update the `## State Check` section with the Step 0 audit command, task/artifact scope, key result, and uncovered parts; for normal success, do not copy the full directory listing or `task.md` tail, and include decisive raw lines only for failures, blocking, identity mismatches, or disputes, before `## Activity Log`
 - Mark all workflow steps as complete
 - Verify and check off all items in `## Completion Checklist` (change `- [ ]` to `- [x]`)
 
@@ -167,36 +171,19 @@ If a summary must mirror a human-decided `post-review-commit` exemption, make a
 
 `--force` cannot lift identity, concurrency, local atomicity, or required-PR hard gates; failures there must stop. Other review, manual-validation, and platform evidence is recorded by terminal verification as warning/pending and can be repaired by retrying.
 
-### 6. Run the Host Finalization Entry Point and Verify the Terminal State
+### 6. Run the Host Finalization Entry Point
 
 ```bash
 agent-infra-internal task-finalization {task-id} complete --agent {standard-agent-token}
 ```
 
-Finalization runs lifecycle -> the terminal task comment -> the `complete-task.completed` gate in a fixed order and records each step in the host receipt. `result=completed` means lifecycle completed safely; if peripheral warnings remain, return `result=completed_with_warnings`, warnings, and pending steps. Use `result=failed` or `result=blocked` only for hard or receipt/capability failures, then fix the cause and retry through the same entry point; do not claim completion or hand-repair partial state.
-
-```bash
-ls .agents/workspace/completed/{task-id}/task.md
-```
-
-Check the task directory only after finalization returns `status=completed`.
+Finalization runs lifecycle -> the terminal task comment -> the `complete-task.completed` gate in a fixed order and records each step in the host receipt. `result=completed` means the host safely completed the task from structured results and the receipt; if peripheral warnings remain, return `result=completed_with_warnings`, warnings, and pending steps. Use `result=failed` or `result=blocked` only for hard or receipt/capability failures, then fix the cause and retry through the same entry point; do not claim completion or hand-repair partial state. A sandbox must not run `ls completed` or a local terminal verification against its historical mount to re-decide this result.
 
 ### 7. Handle Finalization Retries and Results
 
 Both Scenario A and Scenario B `finalization-retry` run the same `task-finalization` entry point from the host. The receipt is only a re-entry hint, not canonical truth: every re-entry revalidates the terminal task comment and completion gate; only an already `completed` task with its short-id registry entry released may skip the irreversible lifecycle. Do not split the operation into the former lifecycle, comment-sync, or completion-gate commands. If the task comment or gate is blocked by the network, preserve the receipt and completed steps, fix the network, and rerun complete-task. If lifecycle has not completed, keep the task active and resume the pending steps from the receipt.
 
-The result must include the structured output from this finalization run, confirming that the task artifact and sync state are valid:
-
-```bash
-agent-infra-internal task-verify {task-id} complete-task.completed --format text
-```
-
-Handle the result as follows:
-- `status=completed` / exit code 0 (all checks passed) -> continue to the "Inform User" step
-- `status=failed` / exit code 1 -> fix the reported issues and run finalization again
-- `status=blocked` / exit code 2 -> preserve the receipt and completed steps and stop; rerun complete-task later to enter `finalization-retry`
-
-Keep the gate output in your reply as fresh evidence. Do not claim completion without output from this run.
+Consume the structured finalization result, receipt, and warning projection directly. Continue only for `completed` or `completed_with_warnings`; preserve the receipt and stop for `failed`, `blocked`, or `unknown`, then retry through the same finalization entry point. Do not run `ls completed` or `task-verify complete-task.completed` from a stale sandbox mount, and do not let that local result overrule the host result.
 
 ### Accepted sandbox-control result recovery
 

package/templates/.agents/skills/complete-task/SKILL.zh-CN.md

@@ -14,6 +14,10 @@ description: >
 
 ## 行为边界 / 关键规则
 
+### 持久化报告证据
+
+生成收尾报告或同步内容时，先读取 `.agents/rules/evidence-reporting.md`。成功检查记录命令、范围、状态/结构化结果、实际结果和未覆盖部分；失败、阻塞或争议保留复现入口、准确位置和决定性摘录。
+
 - 本命令更新任务元数据并物理移动任务目录
 - 除非强制执行，不要转移有未完成工作流步骤的任务
 - 入口接受可选 `--external-pr <N>`；仅用于外部交付候选歧义时的显式选择，不绕过身份或平台门禁
@@ -24,7 +28,7 @@ description: >
 
 在加载 workflow / skill / rules 指令之后、做任何任务状态判断或用户可见结论之前，必须先执行状态核对。指令类文件读取不算对外动作或结论。
 
-运行以下命令，并把原文粘贴到本轮产物的 `## 状态核对` 段：
+运行以下命令，并在本轮产物的 `## 状态核对` 段记录任务/产物范围、关键结果和未覆盖部分；正常成功不粘贴完整目录清单或 `task.md` 尾部。失败、阻塞、身份不一致或争议时，附决定性原文行：
 
 ```bash
 agent-infra-internal task-snapshot {task-id} --format text
@@ -99,7 +103,7 @@ agent-infra-internal platform-pr resolve-external {task-id} --agent {standard-ag
 生命周期前只验证硬门禁：
 - [ ] task 身份、active 状态、并发锁和本地原子生命周期操作可用
 - [ ] `prFlow=required` 时已满足 required-PR delivery
-- [ ] 其余业务证据（工作流、审查、提交、测试、分歧账本、人工校验和平台同步）已记录或可在生命周期后校验
+- [ ] 其余业务证据（工作流、审查、提交、测试、分歧账本、人工校验和平台同步）已记录或可在生命周期后校验；人工验证只能由 committed transaction、receipt、当前 artifact 摘要和带 identity 的通过日志满足
 
 > **⚠️ 前置条件分支判断 — 你必须先判断“继续”还是“停止”：**
 >
@@ -122,7 +126,7 @@ Please satisfy the hard prerequisite first, then retry complete-task.
 ### 3. 完成业务内容更新
 
 在 `.agents/workspace/active/{task-id}/task.md` 中只更新生命周期核心不负责的业务内容：
-- 新增或更新 `## 状态核对` 段，粘贴第 0 步审计命令原文（含 `$ ` 前缀行），放在 `## 活动日志` 之前
+- 新增或更新 `## 状态核对` 段，记录第 0 步审计命令、任务/产物范围、关键结果和未覆盖部分；正常成功不复制完整目录清单或 `task.md` 尾部，失败、阻塞、身份不一致或争议时附决定性原文行，放在 `## 活动日志` 之前
 - 标记所有工作流步骤为已完成
 - 逐项验证并勾选 `## 完成检查清单` 中的所有条目（将 `- [ ]` 改为 `- [x]`）
 
@@ -166,36 +170,19 @@ agent-infra-internal task-verify {task-id} complete-task.preflight --format text
 
 `--force` 不能解除身份、并发、本地原子性或 required-PR hard gate；这些能力失败时必须停止。其余审查、人工校验和平台证据由 terminal verification 记录为 warning/pending，并通过后续重试恢复。
 
-### 6. 执行宿主 finalization 入口并验证终态
+### 6. 执行宿主 finalization 入口
 
 ```bash
 agent-infra-internal task-finalization {task-id} complete --agent {standard-agent-token}
 ```
 
-finalization 按 lifecycle → task 评论 → `complete-task.completed` 校验的固定顺序执行，并将每一步的状态写入宿主 receipt。`result=completed` 即表示 lifecycle 已安全完成；若还有外围 warning，返回 `result=completed_with_warnings`、warnings 和 pending steps。`result=failed` 或 `result=blocked` 仅用于硬失败或 receipt/capability 失败，修复原因后以同一入口重试，不得宣称完成或手工补写局部状态。
-
-```bash
-ls .agents/workspace/completed/{task-id}/task.md
-```
-
-仅在 finalization 返回 `status=completed` 后确认任务目录已成功移动。
+finalization 按 lifecycle → task 评论 → `complete-task.completed` 校验的固定顺序执行，并将每一步的状态写入宿主 receipt。`result=completed` 即表示宿主已依据结构化结果和 receipt 安全完成；若还有外围 warning，返回 `result=completed_with_warnings`、warnings 和 pending steps。`result=failed` 或 `result=blocked` 仅用于硬失败或 receipt/capability 失败，修复原因后以同一入口重试，不得宣称完成或手工补写局部状态。沙箱不得从旧挂载执行 `ls completed` 或本地终态校验来重新裁决该结果。
 
 ### 7. 处理 finalization 重试与结果
 
 场景 A 与场景 B `finalization-retry` 都从宿主执行同一个 `task-finalization` 入口。receipt 只是重入提示，不是 canonical truth：每次重入都要重新验证终态 task 评论和完成校验；只有任务已处于 `completed` 且短号 registry 已释放时，才可跳过不可逆的 lifecycle。不得拆开调用旧的 lifecycle、评论同步或完成校验命令。若 task 评论或校验因网络问题返回 `blocked`，保留 receipt 和已完成状态，修复网络后重跑 complete-task；若生命周期仍未完成，任务保持 active 并从 receipt 的待处理步骤继续。
 
-完成结果必须包含本次 finalization 的结构化输出，确认任务产物和同步状态符合规范：
-
-```bash
-agent-infra-internal task-verify {task-id} complete-task.completed --format text
-```
-
-处理结果：
-- `status=completed` / 退出码 0（全部通过）-> 继续到「告知用户」步骤
-- `status=failed` / 退出码 1 -> 根据输出修复问题后重新运行 finalization
-- `status=blocked` / 退出码 2 -> 保留 receipt 与已完成的步骤并停止；稍后重跑 complete-task 进入 `finalization-retry`
-
-将校验输出保留在回复中作为当次验证输出。没有当次校验输出，不得声明完成。
+完成结果必须直接消费本次宿主 finalization 的结构化输出、receipt 和 warning projection。`completed` 或 `completed_with_warnings` 才允许继续；`failed` / `blocked` / `unknown` 必须保留 receipt 并停止，之后通过同一 finalization 入口重试。不要在沙箱旧挂载中另行运行 `ls completed` 或 `task-verify complete-task.completed`，也不要用其结果推翻宿主结果。
 
 ### accepted 后的 sandbox-control 结果恢复
 

package/templates/.agents/skills/create-task/SKILL.en.md

@@ -126,6 +126,14 @@ date "+%Y-%m-%d %H:%M:%S%z" | sed 's/\([+-][0-9][0-9]\)\([0-9][0-9]\)$/\1:\2/'
 - `blocked`: retain the candidate and retry the same file after the transient problem is fixed.
 - `failed`: report the stable error code. Never modify and retry a candidate under the same idempotency key.
 
+A controlled invocation may return `control` request evidence: `accepted: true`
+with `recovery: none` means the current result can be consumed; `accepted: false`
+with `recovery: new-request-id` means the request was not admitted and the same
+candidate may be retried with a new outer request ID after the transient problem
+is fixed; `recovery: same-request-id` or `recovery: inspect-domain-state` means
+the request was accepted, so do not create a new candidate or replay it
+automatically. Recover with the original request or inspect host task state first.
+
 ### 4. Platform Failure Compatibility
 
 The host service owns platform cascading. The compatibility recovery commands remain visible in the host warning and are not run from the sandbox:

package/templates/.agents/skills/create-task/SKILL.zh-CN.md

@@ -125,6 +125,13 @@ date "+%Y-%m-%d %H:%M:%S%z" | sed 's/\([+-][0-9][0-9]\)\([0-9][0-9]\)$/\1:\2/'
 - `blocked`：保留 candidate 文件，修复暂态问题后用同一文件重试。
 - `failed`：报告稳定错误码；幂等冲突时不得改写原 key 或原 candidate 后重试。
 
+受控调用可能在结果中返回 `control` 请求证据：`accepted: true` 且
+`recovery: none` 表示可以消费当前结果；`accepted: false` 且
+`recovery: new-request-id` 表示请求尚未受理，暂态问题恢复后才能用同一份
+candidate 和新的 outer request ID 重试；`recovery: same-request-id` 或
+`recovery: inspect-domain-state` 表示请求已经受理，不得创建新的 candidate 或
+自动重放，必须先使用原 request 检查结果或宿主任务状态。
+
 ### 4. 平台失败兼容说明
 
 平台级联由宿主服务完成。兼容的人工恢复命令仍由宿主 warning 指引，不在沙箱内执行：

package/templates/.agents/skills/import-codescan/SKILL.en.md

@@ -34,7 +34,7 @@ This skill **creates** task.md, so there is no file to write at the start. Captu
 
 ### 1. Retrieve Alert Information
 
-Read `.agents/rules/security-alerts.md` before this step, then run `bash .agents/scripts/security-alerts.sh read-codescan --number {alert-number}` and parse its JSON result to fetch the alert details.
+Read `.agents/rules/security-alerts.md` before this step, then run `agent-infra-internal platform-security read --kind code-scanning --number {alert-number}` and parse its JSON result to fetch the alert details.
 
 Extract key information:
 - `number`: alert number

package/templates/.agents/skills/import-codescan/SKILL.zh-CN.md

@@ -34,7 +34,7 @@ description: >
 
 ### 1. 获取告警信息
 
-执行前先读取 `.agents/rules/security-alerts.md`，然后运行 `bash .agents/scripts/security-alerts.sh read-codescan --number {alert-number}`，解析其 JSON 结果获取告警详情。
+执行前先读取 `.agents/rules/security-alerts.md`，然后运行 `agent-infra-internal platform-security read --kind code-scanning --number {alert-number}`，解析其 JSON 结果获取告警详情。
 
 提取关键信息：
 - `number`：告警编号

package/templates/.agents/skills/import-dependabot/SKILL.en.md

@@ -34,7 +34,7 @@ This skill **creates** task.md, so there is no file to write at the start. Captu
 
 ### 1. Retrieve Alert Information
 
-Read `.agents/rules/security-alerts.md` before this step, then run `bash .agents/scripts/security-alerts.sh read-dependabot --number {alert-number}` and parse its JSON result to fetch the alert details.
+Read `.agents/rules/security-alerts.md` before this step, then run `agent-infra-internal platform-security read --kind dependabot --number {alert-number}` and parse its JSON result to fetch the alert details.
 
 Extract key information:
 - `number`: alert number

package/templates/.agents/skills/import-dependabot/SKILL.zh-CN.md

@@ -34,7 +34,7 @@ description: >
 
 ### 1. 获取告警信息
 
-执行前先读取 `.agents/rules/security-alerts.md`，然后运行 `bash .agents/scripts/security-alerts.sh read-dependabot --number {alert-number}`，解析其 JSON 结果获取告警详情。
+执行前先读取 `.agents/rules/security-alerts.md`，然后运行 `agent-infra-internal platform-security read --kind dependabot --number {alert-number}`，解析其 JSON 结果获取告警详情。
 
 提取关键信息：
 - `number`：告警编号