d4767ac2e0
* fix: replace /gsd: command format with /gsd- skill format in all suggestions All next-step suggestions shown to users were still using the old colon format (/gsd:xxx) which cannot be copy-pasted as skills. Migrated all occurrences across agents/, commands/, get-shit-done/, docs/, README files, bin/install.js (hardcoded defaults for claude runtime), and get-shit-done/bin/lib/*.cjs (generate-claude-md templates and error messages). Updated tests to assert new hyphen format instead of old colon format. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: migrate remaining /gsd: format to /gsd- in hooks, workflows, and sdk Addresses remaining user-facing occurrences missed in the initial migration: - hooks/: fix 4 user-facing messages (pause-work, update, fast, quick) and 2 comments in gsd-workflow-guard.js - get-shit-done/workflows/: fix 21 Skill() literal calls that Claude executes directly (installer does not transform workflow content) - sdk/prompt-sanitizer.ts: update regex to strip /gsd- format in addition to legacy /gsd: format; update JSDoc comment - tests/: update autonomous-ui-steps, prompt-sanitizer to assert new format Note: commands/gsd/*.md frontmatter (name: gsd:xxx) intentionally unchanged — installer derives skillName from directory path, not the name field. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(plan-phase): preserve --chain flag in auto-advance sync and handle ui-phase gate in chain mode Bug 1: step 15 sync-flag check only guarded against --auto, causing _auto_chain_active to be cleared when plan-phase is invoked without --auto in ARGUMENTS even though a --chain pipeline was active. Added --chain to the guard condition, matching discuss-phase behaviour. Bug 2: UI Design Contract gate (step 5.6) always exited the workflow when UI-SPEC was missing, breaking the discuss --chain pipeline silently. When _auto_chain_active is true, the gate now auto-invokes gsd-ui-phase --auto via Skill() and continues to step 6 without prompting. Manual invocations retain the existing AskUserQuestion flow. * fix: remove <sub>/clear</sub> pattern and duplicate old-format command in discuss-phase.md --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
3.0 KiB
3.0 KiB
模型配置
模型配置控制每个 GSD 代理使用哪个 Claude 模型。这允许平衡质量和 token 消耗。
配置定义
| 代理 | quality |
balanced |
budget |
|---|---|---|---|
| gsd-planner | opus | opus | sonnet |
| gsd-roadmapper | opus | sonnet | sonnet |
| gsd-executor | opus | sonnet | sonnet |
| gsd-phase-researcher | opus | sonnet | haiku |
| gsd-project-researcher | opus | sonnet | haiku |
| gsd-research-synthesizer | sonnet | sonnet | haiku |
| gsd-debugger | opus | sonnet | sonnet |
| gsd-codebase-mapper | sonnet | haiku | haiku |
| gsd-verifier | sonnet | sonnet | haiku |
| gsd-plan-checker | sonnet | sonnet | haiku |
| gsd-integration-checker | sonnet | sonnet | haiku |
| gsd-nyquist-auditor | sonnet | sonnet | haiku |
配置理念
quality - 最大推理能力
- 所有决策代理使用 Opus
- 只读验证使用 Sonnet
- 适用场景：有配额可用、关键架构工作
balanced（默认）- 智能分配
- 仅规划（架构决策发生的地方）使用 Opus
- 执行和研究使用 Sonnet（遵循明确指令）
- 验证使用 Sonnet（需要推理，不仅仅是模式匹配）
- 适用场景：正常开发、质量与成本的良好平衡
budget - 最小化 Opus 使用
- 编写代码的使用 Sonnet
- 研究和验证使用 Haiku
- 适用场景：节省配额、大量工作、不太关键的阶段
解析逻辑
编排器在生成代理前解析模型：
1. 读取 .planning/config.json
2. 检查 model_overrides 是否有代理特定覆盖
3. 如果没有覆盖，在配置表中查找代理
4. 将 model 参数传递给 Task 调用
单代理覆盖
覆盖特定代理而不更改整个配置：
{
"model_profile": "balanced",
"model_overrides": {
"gsd-executor": "opus",
"gsd-planner": "haiku"
}
}
覆盖优先于配置。有效值：opus、sonnet、haiku。
切换配置
运行时：/gsd-set-profile <profile>
项目默认值：在 .planning/config.json 中设置：
{
"model_profile": "balanced"
}
设计理由
为什么 gsd-planner 使用 Opus？ 规划涉及架构决策、目标分解和任务设计。这是模型质量影响最大的地方。
为什么 gsd-executor 使用 Sonnet？ 执行者遵循明确的 PLAN.md 指令。计划已包含推理；执行只是实现。
为什么 balanced 中验证器使用 Sonnet（而非 Haiku）？ 验证需要目标回溯推理 —— 检查代码是否交付了阶段承诺的内容，而不仅仅是模式匹配。Sonnet 处理得很好；Haiku 可能会遗漏细微的差距。
为什么 gsd-codebase-mapper 使用 Haiku？ 只读探索和模式提取。不需要推理，只需从文件内容输出结构化结果。
为什么用 inherit 而不是直接传递 opus？
Claude Code 的 "opus" 别名映射到特定模型版本。组织可能阻止旧版 opus 而允许新版。GSD 为 opus 级代理返回 "inherit"，使其使用用户在会话中配置的任何 opus 版本。这避免了版本冲突和静默回退到 Sonnet。