@fitlab-ai/agent-infra 0.11.2 → 0.11.4

package/templates/.agents/skills/review-analysis/SKILL.en.md

@@ -15,6 +15,10 @@ Review the latest analysis artifact and produce `review-analysis.md` or `review-
 
 ## Boundary / Critical Rules
 
+### Path and rework
+
+This stage consumes only the canonical path fact parsed by core from the latest valid analysis artifact; a stage omitted by that path must not create an artifact. When another Changes Requested verdict follows one completed revision-review cycle in the same stage, classify the rework before the next executor artifact and bind it to the finding, source SHA, and stable task-fact digest. A round or filename change is not new evidence.
+
 ### Persisted Report Evidence
 
 Before generating the review report, read `.agents/rules/evidence-reporting.md`. Normal inspection records the command, scope, structured result, actual conclusion, and uncovered parts; findings, blocking conditions, or disputes retain reproducible locations and decisive excerpts, while identity fields remain exact.
@@ -54,7 +58,7 @@ Require `task.md` and at least one analysis artifact: `analysis.md` or `analysis
 
 ### 2. Resolve Review Context
 
-Run `agent-infra-internal task-artifact {task-id} inspect --family review-analysis`. Continue only for `ready`; take `{analysis-artifact}` from `inputs` and `{review-round}` / `{review-artifact}` from `next.round` / `next.name`. Do not scan rounds or construct names in the skill.
+Run `agent-infra-internal task-artifact {task-id} inspect --family review-analysis`. Continue only for `ready`. When `selection.disposition` is `reuse`, reuse `selection.artifact`, do not run started/init or write a new artifact, and continue directly to completion verification and next-step guidance. Otherwise take `{analysis-artifact}` from `inputs` and `{review-round}` / `{review-artifact}` from `next.round` / `next.name`. Do not scan rounds or construct names in the skill.
 
 ### 3. Read Analysis Context
 

package/templates/.agents/skills/review-analysis/SKILL.zh-CN.md

@@ -15,6 +15,10 @@ description: >
 
 ## 行为边界 / 关键规则
 
+### 路径与返工
+
+本阶段只消费核心从最新有效分析产物解析出的规范路径事实；未被所选路径包含的阶段不得创建产物。同阶段完成一次修订—复审后再次收到 Changes Requested 时，必须在下一执行产物前提交绑定 finding、来源 SHA 和稳定任务事实摘要的返工分类；轮次或文件名变化不是新证据。
+
 ### 持久化报告证据
 
 生成审查报告时，先读取 `.agents/rules/evidence-reporting.md`。正常检视记录命令、范围、结构化结果、实际结论和未覆盖部分；finding、阻塞或争议保留可复现位置与决定性摘录，身份字段必须精确保留。
@@ -57,7 +61,7 @@ agent-infra-internal task-snapshot {task-id} --format text
 
 ### 2. 解析审查上下文
 
-运行 `agent-infra-internal task-artifact {task-id} inspect --family review-analysis`。仅当结果为 `ready` 时继续；从 `inputs` 取得 `{analysis-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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 review-analysis`。仅当结果为 `ready` 时继续。若 `selection.disposition` 为 `reuse`，复用 `selection.artifact`，不得执行 started、init 或写入新产物，并直接进入完成校验与下一步提示。其他状态从 `inputs` 取得 `{analysis-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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。
 
 ### 3. 阅读分析上下文
 

package/templates/.agents/skills/review-code/SKILL.en.md

@@ -15,6 +15,10 @@ Review the latest code round and produce `review-code.md` or `review-code-r{N}.m
 
 ## Boundary / Critical Rules
 
+### Path and rework
+
+This stage consumes only the canonical path fact parsed by core from the latest valid analysis artifact; a stage omitted by that path must not create an artifact. When another Changes Requested verdict follows one completed revision-review cycle in the same stage, classify the rework before the next executor artifact and bind it to the finding, source SHA, and stable task-fact digest. A round or filename change is not new evidence.
+
 ### Persisted Report Evidence
 
 Before generating the review report, read `.agents/rules/evidence-reporting.md`. Normal inspection records the command, scope, structured result, actual conclusion, and uncovered parts; findings, blocking conditions, or disputes retain reproducible locations and decisive excerpts, while identity fields remain exact.
@@ -67,7 +71,7 @@ Require:
 
 ### 2. Resolve the Artifact Context
 
-Run `agent-infra-internal task-artifact {task-id} inspect --family review-code`. Continue only for `ready`; take the latest `{code-artifact}` from `inputs` and `{review-round}` / `{review-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 review-code`. Continue only for `ready`. When `selection.disposition` is `reuse`, reuse `selection.artifact`, do not run started/init or write a new artifact, and continue directly to completion verification and next-step guidance. Otherwise take the latest `{code-artifact}` from `inputs` and `{review-round}` / `{review-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.
 
 ### 3. Read Implementation and Refinement Context
 

package/templates/.agents/skills/review-code/SKILL.zh-CN.md

@@ -15,6 +15,10 @@ description: >
 
 ## 行为边界 / 关键规则
 
+### 路径与返工
+
+本阶段只消费核心从最新有效分析产物解析出的规范路径事实；未被所选路径包含的阶段不得创建产物。同阶段完成一次修订—复审后再次收到 Changes Requested 时，必须在下一执行产物前提交绑定 finding、来源 SHA 和稳定任务事实摘要的返工分类；轮次或文件名变化不是新证据。
+
 ### 持久化报告证据
 
 生成审查报告时，先读取 `.agents/rules/evidence-reporting.md`。正常检视记录命令、范围、结构化结果、实际结论和未覆盖部分；finding、阻塞或争议保留可复现位置与决定性摘录，身份字段必须精确保留。
@@ -66,7 +70,7 @@ agent-infra-internal task-snapshot {task-id} --format text
 
 ### 2. 解析审查上下文
 
-运行 `agent-infra-internal task-artifact {task-id} inspect --family review-code`。仅当结果为 `ready` 时继续；从 `inputs` 取得最新 `{code-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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 review-code`。仅当结果为 `ready` 时继续。若 `selection.disposition` 为 `reuse`，复用 `selection.artifact`，不得执行 started、init 或写入新产物，并直接进入完成校验与下一步提示。其他状态从 `inputs` 取得最新 `{code-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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。
 
 ### 3. 阅读实现与修复上下文
 

package/templates/.agents/skills/review-plan/SKILL.en.md

@@ -15,6 +15,10 @@ Review the latest plan artifact and produce `review-plan.md` or `review-plan-r{N
 
 ## Boundary / Critical Rules
 
+### Path and rework
+
+This stage consumes only the canonical path fact parsed by core from the latest valid analysis artifact; a stage omitted by that path must not create an artifact. When another Changes Requested verdict follows one completed revision-review cycle in the same stage, classify the rework before the next executor artifact and bind it to the finding, source SHA, and stable task-fact digest. A round or filename change is not new evidence.
+
 ### Persisted Report Evidence
 
 Before generating the review report, read `.agents/rules/evidence-reporting.md`. Normal inspection records the command, scope, structured result, actual conclusion, and uncovered parts; findings, blocking conditions, or disputes retain reproducible locations and decisive excerpts, while identity fields remain exact.
@@ -54,7 +58,7 @@ Require `task.md` and at least one plan artifact: `plan.md` or `plan-r{N}.md`.
 
 ### 2. Resolve the Artifact Context
 
-Run `agent-infra-internal task-artifact {task-id} inspect --family review-plan`. Continue only for `ready`; take `{plan-artifact}` from `inputs` and `{review-round}` / `{review-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 review-plan`. Continue only for `ready`. When `selection.disposition` is `reuse`, reuse `selection.artifact`, do not run started/init or write a new artifact, and continue directly to completion verification and next-step guidance. Otherwise take `{plan-artifact}` from `inputs` and `{review-round}` / `{review-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.
 
 ### 3. Read Plan Context
 

package/templates/.agents/skills/review-plan/SKILL.zh-CN.md

@@ -15,6 +15,10 @@ description: >
 
 ## 行为边界 / 关键规则
 
+### 路径与返工
+
+本阶段只消费核心从最新有效分析产物解析出的规范路径事实；未被所选路径包含的阶段不得创建产物。同阶段完成一次修订—复审后再次收到 Changes Requested 时，必须在下一执行产物前提交绑定 finding、来源 SHA 和稳定任务事实摘要的返工分类；轮次或文件名变化不是新证据。
+
 ### 持久化报告证据
 
 生成审查报告时，先读取 `.agents/rules/evidence-reporting.md`。正常检视记录命令、范围、结构化结果、实际结论和未覆盖部分；finding、阻塞或争议保留可复现位置与决定性摘录，身份字段必须精确保留。
@@ -57,7 +61,7 @@ agent-infra-internal task-snapshot {task-id} --format text
 
 ### 2. 解析审查上下文
 
-运行 `agent-infra-internal task-artifact {task-id} inspect --family review-plan`。仅当结果为 `ready` 时继续；从 `inputs` 取得 `{plan-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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 review-plan`。仅当结果为 `ready` 时继续。若 `selection.disposition` 为 `reuse`，复用 `selection.artifact`，不得执行 started、init 或写入新产物，并直接进入完成校验与下一步提示。其他状态从 `inputs` 取得 `{plan-artifact}`，从 `next.round` / `next.name` 取得 `{review-round}` / `{review-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。
 
 ### 3. 阅读方案上下文
 

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

@@ -26,7 +26,7 @@ Before generating a validation report, read `.agents/rules/evidence-reporting.md
 - `complete-manual-validation` remains the maintainer's final registration after coverage is judged sufficient.
 - Never manipulate temporary worktrees, leases, or containers directly; only call `agent-infra-internal task-validate`.
 - Never record tokens, environment variables, full argv, absolute user paths, or raw transcripts in the artifact; the core must write one current-only evidence envelope for each validation target.
-- Before generating a validation artifact Markdown file 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.
+- Before generating validation artifact Markdown referenced by the task comment, read `.agents/rules/sync-content-generation.md` and follow its generator-side constraints. Keep the validation artifact local instead of publishing it as an Issue artifact comment.
 - The branch-only fallback writes to `.agents/workspace/validations/{branch-slug}/` and is marked `recoverable: false`: no task.md write-back, no lifecycle events, no Issue sync, no `task-verify`; the artifact must be carried back to the host that owns the task workspace for registration.
 
 ## Step 0: State Check (pre-execution hard gate)
@@ -46,7 +46,7 @@ Branch-only has no `{task-id}`: skip this step and record `not-applicable (branc
 3. Take the round and artifact name from the core result. After confirming valid explicit work or a non-empty discovered list, run `agent-infra-internal task-event {task-id} validation-run.started --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code}` and classify each item as `executable|unavailable|unknown|unsafe|unresolved`.
 4. Allocate an `{evidence-file}` for each validation target and invoke every executable item separately with `agent-infra-internal task-validate {task-ref} --scope snapshot --format json --evidence-file {evidence-file} -- {command...}`. Only make a second explicit inplace invocation for that item when evidence proves it is required, using its evidence file. If no item is executable, run no fabricated command but still produce coverage-gap evidence.
 5. Read `reference/report-template.md`, create `validation-run.md|validation-run-r{N}.md`, and record the input mode, discovered list, per-item results, CLI JSON allowlist, and sanitized summaries.
-6. Run `agent-infra-internal task-event {task-id} validation-run.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {artifact}`. When an Issue exists, run `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}` and then `agent-infra-internal platform-comment sync {task-id} --kind artifact --artifact {artifact} --agent {standard-agent-token}`.
+6. Run `agent-infra-internal task-event {task-id} validation-run.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {artifact}`. When an Issue exists, run only `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}`; do not publish the validation artifact as an Issue comment.
 7. Run `agent-infra-internal task-verify {task-id} validation-run.completed --artifact {artifact} --format text`; fix failures and rerun it.
 8. Report the evidence path, coverage gaps, and verification result; explicitly leave the decision to run `complete-manual-validation` to the maintainer. Read `.agents/rules/next-step-output.md` and end with `Completed at`.
 

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

@@ -26,7 +26,7 @@ description: >
 - `complete-manual-validation` 仍是维护者确认覆盖充分后的最终登记入口。
 - 禁止直接操作临时 worktree、lease 或 container；只调用 `agent-infra-internal task-validate`。
 - 产物不得记录 token、环境变量、完整 argv、绝对用户路径或原始 transcript；每个验证目标必须由核心写入同一 current-only evidence envelope。
-- 生成会同步到 Issue 的验证 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。
+- 生成会由 task 评论引用的验证 artifact Markdown 前，先读取 `.agents/rules/sync-content-generation.md` 并遵循其中的生成端约束；验证 artifact 保留在本地，不发布为 Issue artifact 评论。
 - branch-only 降级路径走 `.agents/workspace/validations/{branch-slug}/`，标记 `recoverable: false`：不写 task.md、不发 lifecycle 事件、不同步 Issue、不跑 `task-verify`；产物须带回宿主机，由维护者登记。
 
 ## 第 0 步：状态核对（执行前硬约束）
@@ -46,7 +46,7 @@ branch-only 无 `{task-id}`，跳过本步，并在产物 `## 状态核对` 记
 3. 从核心结果取得轮次和产物名；确认存在有效显式工作或非空发现清单后，运行 `agent-infra-internal task-event {task-id} validation-run.started --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code}`，并逐项分类为 `executable|unavailable|unknown|unsafe|unresolved`。
 4. 为每个验证目标分配 `{evidence-file}`，分别调用 `agent-infra-internal task-validate {task-ref} --scope snapshot --format json --evidence-file {evidence-file} -- {command...}`；只有证据表明必须原位时才对该项进行第二次显式 inplace 调用并写入对应 evidence 文件。零项可执行时不运行伪造命令，但仍继续产出覆盖缺口证据。
 5. 读取 `reference/report-template.md`，创建 `validation-run.md|validation-run-r{N}.md`；记录输入模式、发现清单、逐项结果、CLI JSON allowlist 与去敏摘要。
-6. 运行 `agent-infra-internal task-event {task-id} validation-run.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {artifact}`。存在 Issue 时依次运行 `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}` 和 `agent-infra-internal platform-comment sync {task-id} --kind artifact --artifact {artifact} --agent {standard-agent-token}`。
+6. 运行 `agent-infra-internal task-event {task-id} validation-run.completed --agent {standard-agent-token} --initiator {trigger-initiator} --request-id {request-id} --reason-code {reason-code} --artifact {artifact}`。存在 Issue 时仅运行 `agent-infra-internal platform-comment sync {task-id} --kind task --agent {standard-agent-token}`；验证 artifact 不发布为 Issue 评论。
 7. 运行 `agent-infra-internal task-verify {task-id} validation-run.completed --artifact {artifact} --format text`；未通过则修复后重跑。
 8. 告知用户证据路径、覆盖缺口和验证结果；明确仍需维护者判断是否执行 `complete-manual-validation`。读取 `.agents/rules/next-step-output.md`，最后一行输出 `Completed at`。
 

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

@@ -14,14 +14,15 @@ The entry point may omit the task ref; explicit task scope accepts only `--task
 The orchestrator delegates only and never executes a stage skill itself. First read `.agents/rules/no-mid-flow-questions.md`, `.agents/rules/lifecycle-orchestration.md`, and `reference/host-validation.md`.
 
 1. Resolve the canonical task ID, current Agent Client, and optional atomic policy flags, then run `task-snapshot`. If one explicit role field is present, all four are required.
-2. Use the current host's native child spawn, wait, and result interfaces. Task-bound sandboxes forward task operations through the existing broker; retain task binding and checks for real external process operations.
-3. Run `task-orchestration {task-id} begin-or-resume --client {client}` with the complete explicit policy. With no explicit policy, core reads that client's configured policy; existing runs use persisted policy. Disk state outside the complete current structure fails closed without rewriting; finish or clear active runs before upgrading. Stop on paused/completed results.
-4. Run `route`. On `completed`, run `agent-infra-internal task-verify {task-id} run-task.completed --format text` and stop. Only a `running` result with non-null `next` supplies action, role, round, artifact, requested model, and requested effort.
-5. Call `prepare --client {client} --requested-model {requestedModel} --requested-reasoning-effort {requestedReasoningEffort}` with route's exact policy. Codex runs current host preflight. Report actual task, policy, or host failures.
-6. After prepare, run `task-orchestration <task-ref> dispatch` before spawning a fresh native child with route's model/effort. Pass the short task ref, skill, `--orchestrated`, and stage identity. The child calls `await-activation --stage ... --round ... --artifact ... --role ...` to associate its actual identity before executing. Code-task creates the local checkpoint; do not delegate a separate commit stage or push. Retain existing write locks without introducing a new lock protocol around native spawn.
-7. Record actual start and terminal results through SubagentStart/Stop or parent spawn/wait and App Server observations. A timed-out wait does not imply success. Pause on actual identity, transport, or terminal failures. After the stage and child finish, call `advance`; repeat step 4 only while running.
-8. Create a fresh child each round and never reuse a reviewer. Record actual model/effort and host fallback reasons. Validate current identity, artifacts, ledger, and execution results. User rulings override derived historical plan requirements. Do not introduce child discovery, orphan recovery, or a dedicated recovery protocol.
-9. On pause or completion, run the matching typed verification and report the structured endpoint.
+2. Use the current host's native child spawn, wait, and result interfaces. Task-bound sandboxes execute ordinary task orchestration locally in the current worktree; Codex controller `open`, `verify`, and `close` remain restricted broker operations. Retain task binding and checks for real external process operations.
+3. Before each loop, run `agent-infra-internal task-lifecycle {task-id} recover-started --agent {client} --auto`. The client adapter owns recovery capability and stop-evidence validation; an unsupported adapter or a run that needs no recovery returns `no-op`. Continue only after `no-op` or `applied`. Stop on every other result; a later operator invocation retries from persisted state without routing or dispatching a child.
+4. Run `task-orchestration {task-id} begin-or-resume --client {client}` with the complete explicit policy. With no explicit policy, core reads that client's configured policy; existing runs use persisted policy. Disk state outside the complete current structure fails closed without rewriting; finish or clear active runs before upgrading. Stop on paused/completed results.
+5. Run `route`. On `completed`, run `agent-infra-internal task-verify {task-id} run-task.completed --format text` and stop. Only a `running` result with non-null `next` supplies action, role, round, artifact, requested model, and requested effort.
+6. Call `prepare --client {client} --requested-model {requestedModel} --requested-reasoning-effort {requestedReasoningEffort}` with route's exact policy. The client adapter runs host preflight. Report actual task, policy, or host failures.
+7. After prepare, run `task-orchestration <task-ref> dispatch` before spawning a fresh native child with route's model/effort. Pass the short task ref, skill, `--orchestrated`, and stage identity. The child calls `await-activation --stage ... --round ... --artifact ... --role ...` to associate its actual identity before executing. Code-task creates the local checkpoint; do not delegate a separate commit stage or push. Retain existing write locks without introducing a new lock protocol around native spawn.
+8. Record actual start and terminal results through the current host's native lifecycle events and result interfaces. A timed-out wait does not imply success. Pause on actual identity, transport, or terminal failures. After the stage and child finish, call `advance`; repeat step 3 only while running.
+9. Create a fresh child each round and never reuse a reviewer. Record actual model/effort and host fallback reasons. Validate current identity, artifacts, ledger, and execution results. User rulings override derived historical plan requirements. Do not introduce child discovery, orphan recovery, or a dedicated recovery protocol.
+10. On pause or completion, run the matching typed verification and report the structured endpoint.
 
 ## Stop
 

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

@@ -14,14 +14,15 @@ description: >
 总控只编排，不直接执行任何阶段技能。执行前先读取 `.agents/rules/no-mid-flow-questions.md`、`.agents/rules/lifecycle-orchestration.md` 与 `reference/host-validation.md`。
 
 1. 解析规范任务 ID、当前 Agent Client，以及可选的原子策略 `--executor-model`、`--executor-reasoning-effort`、`--reviewer-model`、`--reviewer-reasoning-effort`，并执行 `agent-infra-internal task-snapshot {task-id} --format text`。任一显式策略字段出现时四个 role 字段必须完整，不得与配置拼接；两个角色可以使用同一模型。
-2. 使用当前宿主的原生子 Agent 启动、等待与结果接口。task-bound sandbox 通过现有 broker 转发任务操作；保留任务绑定与真实外部进程操作检查。
-3. 调用 `agent-infra-internal task-orchestration {task-id} begin-or-resume --client {client}` 并转发完整显式策略。完全没有显式策略时由核心读取当前 client 的 `agentClients[].orchestration`；existing run 使用持久化策略。磁盘状态不符合当前完整结构时核心失败关闭且不改写；升级前必须完成或清空 active run。仅当核心返回 `ORCHESTRATION_MODEL_POLICY_REQUIRED` 时，先用 `agent-client model-selection` 展示 complete/partial/interactive-only 来源，再一次收集完整策略；未回答则不创建 run。若为 paused/completed，按结构化结果停止。
-4. 调用 `route` 并读取结构化结果。若返回 `completed`，立即运行 `agent-infra-internal task-verify {task-id} run-task.completed --format text` 并停止；仅当返回 `running` 且 `next` 非空时读取唯一 action、role、round、artifact、`requestedModel` 和 `requestedReasoningEffort`，不得自行推断。
-5. 调用 `prepare --client {client} --requested-model {requestedModel} --requested-reasoning-effort {requestedReasoningEffort}`，使用 route 返回的精确策略。Codex 执行当前宿主 preflight；任务身份、策略或宿主检查失败时记录实际诊断。
-6. prepare 成功后，在调用 fresh 原生子 Agent 前执行 `task-orchestration <task-ref> dispatch`，显式指定 route 返回的 model/effort；传递短任务引用、skill 名、`--orchestrated` 与 stage/round/artifact/role。child 调用 `await-activation --stage ... --round ... --artifact ... --role ...`，确认当前阶段与实际 child 关联后执行。代码阶段通过 code-task 创建本地 checkpoint；禁止单独委派 commit 阶段或推送。保留已有写锁，不把实际 spawn 纳入新建锁协议。
-7. Codex 通过 SubagentStart/Stop 或 parent spawn/wait 与 App Server 记录实际启动及终态。timed-out wait 不推断成功；身份、传输或异常终态失败如实暂停。阶段结果和 child 完成后调用 `advance`；只有 `running` 才重复步骤 4。
-8. 每轮创建新 child；禁止 follow-up 复用 reviewer。如实记录 actual model/effort 及宿主降级理由。检查当前身份、产物、账本与实际执行结果；用户裁决优先于历史方案派生要求。不得新增 child discovery、孤立 child 恢复或专用恢复协议。
-9. 完成或暂停后运行对应 typed verification，并把结构化 run 摘要、暂停原因、commit 终点或 clean completion evidence 告知用户。
+2. 使用当前宿主的原生子 Agent 启动、等待与结果接口。task-bound sandbox 在当前 worktree 本地执行普通任务编排；Codex controller 的 `open`、`verify`、`close` 继续通过受限 broker 执行。保留任务绑定与真实外部进程操作检查。
+3. 每次进入循环时先调用 `agent-infra-internal task-lifecycle {task-id} recover-started --agent {client} --auto`。客户端适配器负责恢复能力和停止证据校验；没有该能力或无需恢复时返回 `no-op`。仅在结构化结果为 `no-op` 或 `applied` 时继续，其他结果安全停止；下一次人为调用按持久化状态幂等重试，不 route、不派发 child。
+4. 调用 `agent-infra-internal task-orchestration {task-id} begin-or-resume --client {client}` 并转发完整显式策略。完全没有显式策略时由核心读取当前 client 的 `agentClients[].orchestration`；existing run 使用持久化策略。磁盘状态不符合当前完整结构时核心失败关闭且不改写；升级前必须完成或清空 active run。仅当核心返回 `ORCHESTRATION_MODEL_POLICY_REQUIRED` 时，先用 `agent-client model-selection` 展示 complete/partial/interactive-only 来源，再一次收集完整策略；未回答则不创建 run。若为 paused/completed，按结构化结果停止。
+5. 调用 `route` 并读取结构化结果。若返回 `completed`，立即运行 `agent-infra-internal task-verify {task-id} run-task.completed --format text` 并停止；仅当返回 `running` 且 `next` 非空时读取唯一 action、role、round、artifact、`requestedModel` 和 `requestedReasoningEffort`，不得自行推断。
+6. 调用 `prepare --client {client} --requested-model {requestedModel} --requested-reasoning-effort {requestedReasoningEffort}`，使用 route 返回的精确策略。客户端适配器执行宿主 preflight；任务身份、策略或宿主检查失败时记录实际诊断。
+7. prepare 成功后，在调用 fresh 原生子 Agent 前执行 `task-orchestration <task-ref> dispatch`，显式指定 route 返回的 model/effort；传递短任务引用、skill 名、`--orchestrated` 与 stage/round/artifact/role。child 调用 `await-activation --stage ... --round ... --artifact ... --role ...`，确认当前阶段与实际 child 关联后执行。代码阶段通过 code-task 创建本地 checkpoint；禁止单独委派 commit 阶段或推送。保留已有写锁，不把实际 spawn 纳入新建锁协议。
+8. 通过当前宿主的原生生命周期事件和结果接口记录实际启动及终态。timed-out wait 不推断成功；身份、传输或异常终态失败如实暂停。阶段结果和 child 完成后调用 `advance`；只有 `running` 才重复步骤 3。
+9. 每轮创建新 child；禁止 follow-up 复用 reviewer。如实记录 actual model/effort 及宿主降级理由。检查当前身份、产物、账本与实际执行结果；用户裁决优先于历史方案派生要求。不得新增 child discovery、孤立 child 恢复或专用恢复协议。
+10. 完成或暂停后运行对应 typed verification，并把结构化 run 摘要、暂停原因、commit 终点或 clean completion evidence 告知用户。
 
 ## 停止
 

package/templates/.agents/skills/update-agent-infra/scripts/sync-templates.js

@@ -363,6 +363,13 @@ function assetMatches(entry, target) {
     : normalizedTarget === normalizedEntry;
 }
 
+function isOwnedByDisabledAgentClient(target, enabledSet) {
+  return AGENT_CLIENT_MANIFEST.some((adapter) =>
+    !enabledSet.has(adapter.id)
+    && adapter.ownedPathPrefixes.some((prefix) => assetMatches(prefix, target))
+  );
+}
+
 function adapterAssets(adapters, category) {
   return adapters.flatMap((adapter) => adapter[category]);
 }
@@ -1914,6 +1921,15 @@ function syncTemplates(projectRoot, templateRootOverride) {
   }
 
   const mergedMap = new Map();
+  const addMergedTarget = (target, source) => {
+    if (
+      isOwnedByDisabledAgentClient(target, enabledTUIs)
+      && !isCustomProtected(target, protectedCustomSkills, customCommandTargets)
+    ) {
+      return;
+    }
+    if (!mergedMap.has(target)) mergedMap.set(target, source);
+  };
   for (const entry of merged) {
     if (isPathOwnedByOtherPlatform(entry, platformType)) {
       report.managed.skippedPlatform.push(entry);
@@ -1925,13 +1941,13 @@ function syncTemplates(projectRoot, templateRootOverride) {
         return globMatch(entry, t);
       });
       for (const [t, s] of platformSelect(langSelect(hits, lang, allSet, project), platformType, project)) {
-        if (!mergedMap.has(t)) mergedMap.set(t, s);
+        addMergedTarget(t, s);
       }
     } else {
       const rels = entryVariantRels(entry, allSet, platformType);
       const selected = platformSelect(langSelect(rels, lang, allSet, project), platformType, project);
       for (const [t, s] of selected) {
-        if (!mergedMap.has(t)) mergedMap.set(t, s);
+        addMergedTarget(t, s);
       }
     }
   }

package/templates/.agents/templates/task.en.md

@@ -106,8 +106,8 @@ delivery_remote_head:          # Most recent successfully delivered task branch
 
 ## Rework Intent
 
-| intent_id | finding_id | source_artifact | source_sha256 | target | status | declared_at | consumed_at |
-|-----------|------------|----------------|---------------|--------|--------|-------------|-------------|
+| intent_id | finding_id | source_artifact | source_sha256 | target | classification | evidence_digest | task_fact_digest | status | declared_at | consumed_at |
+|-----------|------------|----------------|---------------|--------|----------------|-----------------|------------------|--------|-------------|-------------|
 
 ## Artifact Invalidation
 

package/templates/.agents/templates/task.zh-CN.md

@@ -106,8 +106,8 @@ delivery_remote_head:          # 最近一次成功交付到 remote 的任务分
 
 ## 返工意图
 
-| intent_id | finding_id | source_artifact | source_sha256 | target | status | declared_at | consumed_at |
-|-----------|------------|----------------|---------------|--------|--------|-------------|-------------|
+| intent_id | finding_id | source_artifact | source_sha256 | target | classification | evidence_digest | task_fact_digest | status | declared_at | consumed_at |
+|-----------|------------|----------------|---------------|--------|----------------|-----------------|------------------|--------|-------------|-------------|
 
 ## 产物失效记录
 

package/templates/.agents/workflows/bug-fix.en.yaml

@@ -9,6 +9,11 @@
 name: bug-fix
 description: Workflow for diagnosing and fixing a defect.
 
+lifecycle_paths:
+  streamlined: [analysis, code, code-review]
+  standard: [analysis, design, code, code-review]
+  full: [analysis, analysis-review, design, design-review, code, code-review]
+
 steps:
   - name: analysis
     description: Reproduce the defect, identify the root cause, and define the fix scope.
@@ -29,7 +34,7 @@ steps:
       outputs:
         - name: analysis
           pattern: "analysis.md | analysis-r{N}.md"
-          rule: "Scan existing analysis artifacts; create analysis.md for round 1 and analysis-r{N}.md for later revisions (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Root-cause analysis
       - Affected files list
@@ -56,7 +61,7 @@ steps:
       outputs:
         - name: review-analysis
           pattern: "review-analysis.md | review-analysis-r{N}.md"
-          rule: "Scan existing requirement-analysis review artifacts; create review-analysis.md for round 1 and review-analysis-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Requirement analysis review report
       - Findings to fix, if any
@@ -84,7 +89,7 @@ steps:
       outputs:
         - name: plan
           pattern: "plan.md | plan-r{N}.md"
-          rule: "Scan existing plan artifacts; create plan.md for round 1 and plan-r{N}.md for later revisions (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Fix plan
       - Regression test list
@@ -110,7 +115,7 @@ steps:
       outputs:
         - name: review-plan
           pattern: "review-plan.md | review-plan-r{N}.md"
-          rule: "Scan existing plan review artifacts; create review-plan.md for round 1 and review-plan-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Plan review report
       - Findings to fix, if any
@@ -146,7 +151,7 @@ steps:
       outputs:
         - name: code
           pattern: "code.md | code-r{N}.md"
-          rule: "Scan existing code artifacts; create code.md for round 1 and code-r{N}.md for later fixes or reimplementation (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Bug-fix branch with implementation or fixes
       - Regression test files
@@ -173,7 +178,7 @@ steps:
       outputs:
         - name: review-code
           pattern: "review-code.md | review-code-r{N}.md"
-          rule: "Scan existing code review artifacts; create review-code.md for round 1 and review-code-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Code review report
       - Findings list, if any

package/templates/.agents/workflows/bug-fix.zh-CN.yaml

@@ -8,6 +8,11 @@
 name: bug-fix
 description: 诊断和修复缺陷的工作流。
 
+lifecycle_paths:
+  streamlined: [analysis, code, code-review]
+  standard: [analysis, design, code, code-review]
+  full: [analysis, analysis-review, design, design-review, code, code-review]
+
 steps:
   - name: analysis
     description: 复现缺陷，识别根本原因，确定修复范围。
@@ -28,7 +33,7 @@ steps:
       outputs:
         - name: analysis
           pattern: "analysis.md | analysis-r{N}.md"
-          rule: "扫描现有分析产物；首轮创建 analysis.md，后续修订创建 analysis-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 根因分析
       - 受影响文件列表
@@ -55,7 +60,7 @@ steps:
       outputs:
         - name: review-analysis
           pattern: "review-analysis.md | review-analysis-r{N}.md"
-          rule: "扫描现有需求分析审查产物；首轮创建 review-analysis.md，后续轮次创建 review-analysis-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 需求分析审查报告
       - 待修正问题列表（如有）
@@ -83,7 +88,7 @@ steps:
       outputs:
         - name: plan
           pattern: "plan.md | plan-r{N}.md"
-          rule: "扫描现有方案产物；首轮创建 plan.md，后续修订创建 plan-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 任务文件中记录的修复方案
       - 待添加的测试用例列表
@@ -109,7 +114,7 @@ steps:
       outputs:
         - name: review-plan
           pattern: "review-plan.md | review-plan-r{N}.md"
-          rule: "扫描现有方案审查产物；首轮创建 review-plan.md，后续轮次创建 review-plan-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 技术方案审查报告
       - 待修正问题列表（如有）
@@ -145,7 +150,7 @@ steps:
       outputs:
         - name: code
           pattern: "code.md | code-r{N}.md"
-          rule: "扫描现有实现产物；首轮创建 code.md，后续修复或重实现创建 code-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 包含实现或修复的缺陷修复分支
       - 回归测试文件
@@ -172,7 +177,7 @@ steps:
       outputs:
         - name: review-code
           pattern: "review-code.md | review-code-r{N}.md"
-          rule: "扫描现有代码审查产物；首轮创建 review-code.md，后续轮次创建 review-code-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 代码审查报告
       - 问题列表（如有）

package/templates/.agents/workflows/feature-development.en.yaml

@@ -9,6 +9,11 @@
 name: feature-development
 description: End-to-end workflow for developing a new feature.
 
+lifecycle_paths:
+  streamlined: [analysis, code, code-review]
+  standard: [analysis, design, code, code-review]
+  full: [analysis, analysis-review, design, design-review, code, code-review]
+
 steps:
   - name: analysis
     description: Analyze feature requirements, explore the codebase, and identify affected areas.
@@ -27,7 +32,7 @@ steps:
       outputs:
         - name: analysis
           pattern: "analysis.md | analysis-r{N}.md"
-          rule: "Scan existing analysis artifacts; create analysis.md for round 1 and analysis-r{N}.md for later revisions (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Requirement analysis report
       - Affected files list
@@ -54,7 +59,7 @@ steps:
       outputs:
         - name: review-analysis
           pattern: "review-analysis.md | review-analysis-r{N}.md"
-          rule: "Scan existing requirement-analysis review artifacts; create review-analysis.md for round 1 and review-analysis-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Requirement analysis review report
       - Findings to fix, if any
@@ -83,7 +88,7 @@ steps:
       outputs:
         - name: plan
           pattern: "plan.md | plan-r{N}.md"
-          rule: "Scan existing plan artifacts; create plan.md for round 1 and plan-r{N}.md for later revisions (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Technical plan
       - Interface definitions
@@ -110,7 +115,7 @@ steps:
       outputs:
         - name: review-plan
           pattern: "review-plan.md | review-plan-r{N}.md"
-          rule: "Scan existing plan review artifacts; create review-plan.md for round 1 and review-plan-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Plan review report
       - Findings to fix, if any
@@ -146,7 +151,7 @@ steps:
       outputs:
         - name: code
           pattern: "code.md | code-r{N}.md"
-          rule: "Scan existing code artifacts; create code.md for round 1 and code-r{N}.md for later fixes or reimplementation (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Feature branch with implementation or fixes
       - Test files
@@ -173,7 +178,7 @@ steps:
       outputs:
         - name: review-code
           pattern: "review-code.md | review-code-r{N}.md"
-          rule: "Scan existing code review artifacts; create review-code.md for round 1 and review-code-r{N}.md for later rounds (N = current highest round + 1)"
+          rule: "Let the task-artifact selector decide: reuse a completed artifact when inputs and substantive results match; create the next round only for new inputs, rework evidence, or result changes"
     outputs:
       - Code review report
       - Findings to fix

package/templates/.agents/workflows/feature-development.zh-CN.yaml

@@ -8,6 +8,11 @@
 name: feature-development
 description: 开发新功能的端到端工作流。
 
+lifecycle_paths:
+  streamlined: [analysis, code, code-review]
+  standard: [analysis, design, code, code-review]
+  full: [analysis, analysis-review, design, design-review, code, code-review]
+
 steps:
   - name: analysis
     description: 分析功能需求，探索代码库，识别受影响的区域。
@@ -26,7 +31,7 @@ steps:
       outputs:
         - name: analysis
           pattern: "analysis.md | analysis-r{N}.md"
-          rule: "扫描现有分析产物；首轮创建 analysis.md，后续修订创建 analysis-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 需求分析报告
       - 受影响文件列表
@@ -53,7 +58,7 @@ steps:
       outputs:
         - name: review-analysis
           pattern: "review-analysis.md | review-analysis-r{N}.md"
-          rule: "扫描现有需求分析审查产物；首轮创建 review-analysis.md，后续轮次创建 review-analysis-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 需求分析审查报告
       - 待修正问题列表（如有）
@@ -82,7 +87,7 @@ steps:
       outputs:
         - name: plan
           pattern: "plan.md | plan-r{N}.md"
-          rule: "扫描现有方案产物；首轮创建 plan.md，后续修订创建 plan-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 技术方案
       - 接口定义
@@ -109,7 +114,7 @@ steps:
       outputs:
         - name: review-plan
           pattern: "review-plan.md | review-plan-r{N}.md"
-          rule: "扫描现有方案审查产物；首轮创建 review-plan.md，后续轮次创建 review-plan-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 技术方案审查报告
       - 待修正问题列表（如有）
@@ -145,7 +150,7 @@ steps:
       outputs:
         - name: code
           pattern: "code.md | code-r{N}.md"
-          rule: "扫描现有实现产物；首轮创建 code.md，后续修复或重实现创建 code-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 包含实现或修复的功能分支
       - 测试文件
@@ -172,7 +177,7 @@ steps:
       outputs:
         - name: review-code
           pattern: "review-code.md | review-code-r{N}.md"
-          rule: "扫描现有代码审查产物；首轮创建 review-code.md，后续轮次创建 review-code-r{N}.md（N = 当前最高轮次 + 1）"
+          rule: "由 task-artifact 选择器决定：输入与实质结果未变化时复用已完成产物；仅在新输入、新返工证据或结果变化时创建下一轮"
     outputs:
       - 代码审查报告
       - 待修复问题列表