gsd-plan-review-convergence

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

<objective> Cross-AI plan convergence loop — an outer revision gate around gsd-review and gsd-planner. Repeatedly: review plans with external AI CLIs → if HIGH or actionable non-HIGH concerns remain → replan with --reviews feedback → re-review. Stops when no unresolved HIGH concerns or actionable MEDIUM/LOW findings remain outside PLAN.md, or when max cycles is reached.

Flow: Skill("gsd-plan-phase") → Agent→Skill("gsd-review") → check unresolved HIGH + actionable non-HIGH → Skill("gsd-plan-phase --reviews") → Agent→Skill("gsd-review") → ... → Converge or escalate

Replaces gsd-plan-phase's internal gsd-plan-checker with external AI reviewers (codex, gemini, etc.). Plan-phase runs inline (bare Skill at depth 0) so it can spawn gsd-planner/gsd-plan-checker at depth 1. Review runs inside an isolated Agent (gsd-review is a Bash leaf — no sub-agents needed). Orchestrator only does loop control.

Orchestrator role: Parse arguments, validate phase, run plan-phase inline (Skill at depth 0), spawn an Agent for gsd-review, check unresolved HIGH and actionable non-HIGH counts, stall detection, escalation gate. </objective>

<execution_context> @$HOME/.claude/gsd-core/workflows/plan-review-convergence.md @$HOME/.claude/gsd-core/references/revision-loop.md @$HOME/.claude/gsd-core/references/gates.md @$HOME/.claude/gsd-core/references/agent-contracts.md </execution_context>

<runtime_note> Copilot (VS Code): Use

vscode_askquestions

wherever this workflow calls

AskUserQuestion

. They are equivalent —

vscode_askquestions

is the VS Code Copilot implementation of the same interactive question API. Do not skip questioning steps because

AskUserQuestion

appears unavailable; use

vscode_askquestions

instead. </runtime_note>

<context> Phase number: extracted from $ARGUMENTS (required)

Flags:

```
--codex
```
— Use Codex CLI as reviewer (default if no reviewer flag given AND
```
review.default_reviewers
```
is unset; otherwise
```
review.default_reviewers
```
wins per ADR-0011 — #2315)
```
--gemini
```
— Use Gemini CLI as reviewer
```
--agy
```
/
```
--antigravity
```
— Use Antigravity CLI as reviewer (successor to the discontinued Gemini CLI)
```
--claude
```
— Use Claude CLI as reviewer (separate session)
```
--coderabbit
```
— Use CodeRabbit as reviewer (reviews the working-tree diff, not the source tree)
```
--opencode
```
— Use OpenCode as reviewer
```
--qwen
```
— Use Qwen Code CLI as reviewer (Alibaba Qwen models)
```
--cursor
```
— Use Cursor agent as reviewer
```
--ollama
```
— Use local Ollama server as reviewer (OpenAI-compatible, default host
```
http://localhost:11434
```
; configure model via
```
review.models.ollama
```
)
```
--lm-studio
```
— Use local LM Studio server as reviewer (OpenAI-compatible, default host
```
http://localhost:1234
```
; configure model via
```
review.models.lm_studio
```
)
```
--llama-cpp
```
— Use local llama.cpp server as reviewer (OpenAI-compatible, default host
```
http://localhost:8080
```
; configure model via
```
review.models.llama_cpp
```
)
```
--kimi-code
```
— Use Kimi Code CLI as reviewer (Moonshot AI)
```
--all
```
— Use all available CLIs and running local model servers
```
--max-cycles N
```
— Maximum replan→review cycles (default: 3)

Feature gate: This command requires

workflow.plan_review_convergence=true

. Enable with:

gsd config-set workflow.plan_review_convergence true

</context>

<process> Execute end-to-end. Preserve all workflow gates (pre-flight, revision loop, stall detection, escalation). </process>

<objective> 跨AI计划收敛循环——围绕gsd-review和gsd-planner的外部修订关卡。重复执行：使用外部AI CLI评审计划→如果仍存在高优先级（HIGH）或可处理的非高优先级问题→通过--reviews反馈重新规划→再次评审。当PLAN.md之外没有未解决的高优先级问题或可处理的中/低优先级发现，或者达到最大循环次数时停止。

流程： Skill("gsd-plan-phase") → Agent→Skill("gsd-review") → 检查未解决的高优先级及可处理的非高优先级问题 → Skill("gsd-plan-phase --reviews") → Agent→Skill("gsd-review") → ... → 收敛或升级

用外部AI评审器（codex、gemini等）替代gsd-plan-phase内部的gsd-plan-checker。Plan-phase运行内联（深度为0的基础Skill），因此它可以在深度1处生成gsd-planner/gsd-plan-checker。评审在独立的Agent中运行（gsd-review是Bash叶子节点——无需子代理）。编排器仅负责循环控制。

编排器角色： 解析参数，验证阶段，内联运行plan-phase（深度为0的Skill），为gsd-review生成Agent，检查未解决的高优先级和可处理的非高优先级问题数量，停滞检测，升级关卡。 </objective>

<runtime_note> Copilot（VS Code）： 在此工作流调用

AskUserQuestion

的任何位置，使用

vscode_askquestions

。二者功能等效——

vscode_askquestions

是VS Code Copilot对同一交互式问题API的实现。不要因为

AskUserQuestion

看似不可用而跳过提问步骤；请改用

vscode_askquestions

。 </runtime_note>

<context> 阶段编号：从$ARGUMENTS中提取（必填）

标志：

```
--codex
```
— 使用Codex CLI作为评审器（如果未指定评审器标志且
```
review.default_reviewers
```
未设置，则为默认选项；否则根据ADR-0011，
```
review.default_reviewers
```
优先——#2315）
```
--gemini
```
— 使用Gemini CLI作为评审器
```
--agy
```
/
```
--antigravity
```
— 使用Antigravity CLI作为评审器（已停用的Gemini CLI的继任者）
```
--claude
```
— 使用Claude CLI作为评审器（独立会话）
```
--coderabbit
```
— 使用CodeRabbit作为评审器（评审工作树差异，而非源树）
```
--opencode
```
— 使用OpenCode作为评审器
```
--qwen
```
— 使用Qwen Code CLI作为评审器（阿里巴巴通义千问模型）
```
--cursor
```
— 使用Cursor agent作为评审器
```
--ollama
```
— 使用本地Ollama服务器作为评审器（兼容OpenAI，默认主机
```
http://localhost:11434
```
；通过
```
review.models.ollama
```
配置模型）
```
--lm-studio
```
— 使用本地LM Studio服务器作为评审器（兼容OpenAI，默认主机
```
http://localhost:1234
```
；通过
```
review.models.lm_studio
```
配置模型）
```
--llama-cpp
```
— 使用本地llama.cpp服务器作为评审器（兼容OpenAI，默认主机
```
http://localhost:8080
```
；通过
```
review.models.llama_cpp
```
配置模型）
```
--kimi-code
```
— 使用Kimi Code CLI作为评审器（Moonshot AI）
```
--all
```
— 使用所有可用的CLI和运行中的本地模型服务器
```
--max-cycles N
```
— 最大重新规划→评审循环次数（默认值：3）

功能开关： 此命令需要

workflow.plan_review_convergence=true

。通过以下命令启用：

gsd config-set workflow.plan_review_convergence true

</context>

<process> 端到端执行。保留所有工作流关卡（预检查、修订循环、停滞检测、升级）。 </process>