@sushanglewis/lincoln 1.5.5 → 1.6.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/main.js +425 -51
- package/dist/main.js.map +1 -1
- package/dist/postinstall.d.ts +9 -2
- package/dist/postinstall.js +812 -6
- package/dist/postinstall.js.map +1 -1
- package/framework/.claude/agents/architect.md +15 -15
- package/framework/.claude/agents/default.md +118 -104
- package/framework/.claude/agents/designer.md +16 -51
- package/framework/.claude/agents/engineer.md +16 -25
- package/framework/.claude/agents/frontend-engineer.md +15 -16
- package/framework/.claude/agents/knowledge-curator.md +18 -15
- package/framework/.claude/agents/pm.md +16 -75
- package/framework/.claude/agents/qa.md +17 -13
- package/framework/.claude/agents/release-coordinator.md +18 -12
- package/framework/.claude/agents/researcher.md +16 -26
- package/framework/.claude/harnesses/command-map.yaml +13 -6
- package/framework/.claude/hooks/on-session-start.sh +0 -1
- package/framework/.claude/schemas/agent-definition.schema.json +6 -1
- package/framework/.claude/schemas/stage-definition.schema.json +49 -4
- package/framework/.claude/skills/build-product-prototype/prompts/main.md +89 -26
- package/framework/.claude/skills/clarify-requirements/prompts/main.md +80 -14
- package/framework/.claude/skills/draft-product-design/prompts/main.md +24 -12
- package/framework/.claude/skills/lc-build-codebase-knowledge/SKILL.md +2 -0
- package/framework/.claude/skills/lc-explore-opensource/SKILL.md +2 -0
- package/framework/.claude/skills/lc-init-branch/SKILL.md +2 -2
- package/framework/.claude/skills/lc-market-research/SKILL.md +1 -1
- package/framework/.claude/skills/lc-one-id/SKILL.md +81 -0
- package/framework/.claude/skills/lc-one-id/prompts/main.md +35 -0
- package/framework/.claude/skills/lc-stakeholder-research/SKILL.md +1 -1
- package/framework/.claude/skills/lc-workflow-router/SKILL.md +2 -0
- package/framework/.claude/stages/build-codebase-knowledge.yaml +10 -10
- package/framework/.claude/stages/clarify.yaml +21 -15
- package/framework/.claude/stages/explore-opensource.yaml +13 -10
- package/framework/.claude/stages/implement.yaml +11 -11
- package/framework/.claude/stages/ingest.yaml +13 -10
- package/framework/.claude/stages/lc-analyze-frameworks.yaml +13 -10
- package/framework/.claude/stages/lc-collect-intelligence.yaml +13 -10
- package/framework/.claude/stages/lc-competitive-analysis.yaml +13 -10
- package/framework/.claude/stages/lc-first-principles.yaml +13 -10
- package/framework/.claude/stages/lc-market-research.yaml +13 -10
- package/framework/.claude/stages/lc-product-research.yaml +13 -10
- package/framework/.claude/stages/lc-research-report.yaml +13 -10
- package/framework/.claude/stages/lc-research-scope.yaml +13 -10
- package/framework/.claude/stages/lc-stakeholder-research.yaml +13 -10
- package/framework/.claude/stages/lc-storytelling.yaml +13 -10
- package/framework/.claude/stages/product-design-docs.yaml +30 -18
- package/framework/.claude/stages/product-prototype.yaml +20 -17
- package/framework/.claude/stages/propose.yaml +16 -9
- package/framework/.claude/stages/split.yaml +13 -9
- package/framework/.claude/stages/sync-knowledge.yaml +20 -9
- package/framework/.claude/stages/tdd-development-plan.yaml +17 -9
- package/framework/.claude/stages/workflow-router.yaml +17 -13
- package/framework/.claude/workflows/bug-fix.yaml +1 -0
- package/framework/.claude/workflows/existing-project-iteration.yaml +1 -0
- package/framework/.claude/workflows/interview-to-knowledge.yaml +1 -0
- package/framework/.claude/workflows/oss-first-design.yaml +1 -0
- package/framework/.claude-plugin/marketplace.json +2 -2
- package/framework/.claude-plugin/plugin.json +2 -1
- package/framework/.version-bump.json +1 -1
- package/framework/README.en.md +15 -2
- package/framework/README.md +16 -3
- package/framework/RELEASE.md +48 -0
- package/framework/manifest.json +94 -74
- package/framework/scripts/check-prompt-drift.py +132 -1
- package/framework/scripts/extract_release_notes.py +79 -0
- package/framework/scripts/init-lincoln-branch.sh +6 -0
- package/framework/scripts/lincoln_command_map.py +1 -1
- package/framework/scripts/lincoln_id.py +333 -0
- package/framework/scripts/lincoln_index.py +10 -0
- package/framework/scripts/lincoln_prd.py +16 -17
- package/framework/scripts/lincoln_render.py +657 -0
- package/framework/scripts/run-infrastructure-tests.py +2 -0
- package/framework/scripts/stage_loader.py +149 -2
- package/framework/scripts/static-check.sh +1 -1
- package/framework/scripts/update_readme_version.py +149 -0
- package/framework/scripts/validate_stage.py +80 -1
- package/framework/tools/lincoln/package-lock.json +2 -2
- package/framework/tools/lincoln/package.json +1 -1
- package/framework/tools/lincoln-installer/package-lock.json +2 -2
- package/framework/tools/lincoln-installer/package.json +1 -1
- package/package.json +2 -2
- package/framework/.claude/agents/_contract.md +0 -79
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: explore-opensource
|
|
3
3
|
name: Explore Open Source
|
|
4
4
|
description: 'Researches existing open-source projects that can inform
|
|
@@ -6,31 +6,34 @@ description: 'Researches existing open-source projects that can inform
|
|
|
6
6
|
or solve the current requirement. Triggered by oss-first-design
|
|
7
7
|
|
|
8
8
|
template after clarify.'
|
|
9
|
-
|
|
9
|
+
workflows:
|
|
10
10
|
- oss-first-design
|
|
11
11
|
prerequisite_stage: clarify
|
|
12
12
|
next_stage: product-design-docs
|
|
13
13
|
human_gate: false
|
|
14
|
+
execution_mode: solo
|
|
15
|
+
primary_action: lc-explore-opensource
|
|
16
|
+
variables:
|
|
17
|
+
- process_slug
|
|
18
|
+
- change_name
|
|
14
19
|
agent:
|
|
15
20
|
primary: lc-researcher
|
|
16
21
|
reviewers:
|
|
17
22
|
- lc-engineer
|
|
18
23
|
handoff_to: lc-pm
|
|
19
24
|
skills:
|
|
20
|
-
required:
|
|
25
|
+
required:
|
|
26
|
+
- lincoln:lc-explore-opensource
|
|
21
27
|
optional: []
|
|
22
28
|
artifacts:
|
|
23
29
|
required:
|
|
24
30
|
- '{process_slug}/docs/research/{change_name}-oss-options.md'
|
|
31
|
+
references:
|
|
32
|
+
skill_prompt: .claude/skills/lc-explore-opensource/SKILL.md
|
|
33
|
+
agent_guide: .claude/agents/researcher.md
|
|
25
34
|
context:
|
|
26
35
|
goal: 在设计阶段探索开源方案,避免从零造轮子,并为实现提供参考。
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
- 禁止修改 {process_slug}/recordings/ 中的原始文件。
|
|
30
|
-
|
|
31
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
32
|
-
|
|
33
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
36
|
+
execution: "调用 lc-explore-opensource skill 调研可复用的开源方案,评估其适配性、许可与维护状况,产出 {process_slug}/docs/research/{change_name}-oss-options.md,为产品设计与实现提供决策参考。"
|
|
34
37
|
gates:
|
|
35
38
|
entry:
|
|
36
39
|
- check: previous_stage_completed
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: implement
|
|
3
3
|
name: Implement
|
|
4
4
|
description: 'Core implementation stage. Engineering executes tasks with TDD
|
|
@@ -8,10 +8,16 @@ description: 'Core implementation stage. Engineering executes tasks with TDD
|
|
|
8
8
|
for human engineering work; test/verify and ship occur within
|
|
9
9
|
|
|
10
10
|
this stage or as follow-up PR lifecycle steps.'
|
|
11
|
-
|
|
11
|
+
workflows:
|
|
12
|
+
- bug-fix
|
|
13
|
+
- existing-project-iteration
|
|
14
|
+
- interview-to-knowledge
|
|
15
|
+
- oss-first-design
|
|
12
16
|
prerequisite_stage: split
|
|
13
17
|
next_stage: sync-knowledge
|
|
14
18
|
human_gate: true
|
|
19
|
+
execution_mode: both
|
|
20
|
+
primary_action: implement
|
|
15
21
|
agent:
|
|
16
22
|
primary: lc-engineer
|
|
17
23
|
reviewers:
|
|
@@ -40,17 +46,11 @@ skills:
|
|
|
40
46
|
- oh-my-claudecode:ultrawork
|
|
41
47
|
artifacts:
|
|
42
48
|
required: []
|
|
49
|
+
references:
|
|
50
|
+
agent_guide: .claude/agents/engineer.md
|
|
43
51
|
context:
|
|
44
52
|
goal: 人类研发团队基于已创建的 GitHub Issues 进行代码开发并提交 PR。
|
|
45
|
-
|
|
46
|
-
|
|
47
|
-
- 禁止修改 {process_slug}/recordings/ 中的原始文件。
|
|
48
|
-
|
|
49
|
-
- 本阶段设有 human_gate,必须获得人类 PM 显式确认后才能进入下一阶段。
|
|
50
|
-
|
|
51
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
52
|
-
|
|
53
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
53
|
+
execution: 人类研发团队基于 split 阶段创建的 GitHub Issues 按 TDD 纪律进行代码开发、评审与 PR 提交;Agent 在此阶段暂停主动驱动,由人类工程师主导实现工作,PR 合并后触发 sync-knowledge。
|
|
54
54
|
gates:
|
|
55
55
|
entry:
|
|
56
56
|
- check: previous_stage_completed
|
|
@@ -1,19 +1,25 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: ingest
|
|
3
3
|
name: Ingest
|
|
4
4
|
description: 'Ingests raw interview recordings and produces transcripts,
|
|
5
5
|
|
|
6
6
|
summaries, and raw insights.'
|
|
7
|
-
|
|
7
|
+
workflows:
|
|
8
8
|
- interview-to-knowledge
|
|
9
9
|
prerequisite_stage: workflow-router
|
|
10
10
|
next_stage: clarify
|
|
11
11
|
human_gate: false
|
|
12
|
+
execution_mode: team
|
|
13
|
+
primary_action: process-interview
|
|
14
|
+
variables:
|
|
15
|
+
- process_slug
|
|
16
|
+
- session_id
|
|
12
17
|
agent:
|
|
13
18
|
primary: lc-pm
|
|
14
19
|
handoff_to: lc-pm
|
|
15
20
|
skills:
|
|
16
|
-
required:
|
|
21
|
+
required:
|
|
22
|
+
- lincoln:process-interview
|
|
17
23
|
optional:
|
|
18
24
|
- oh-my-claudecode:deep-interview
|
|
19
25
|
artifacts:
|
|
@@ -22,15 +28,12 @@ artifacts:
|
|
|
22
28
|
- '{process_slug}/interviews/{session_id}/transcript.md'
|
|
23
29
|
- '{process_slug}/interviews/{session_id}/summary.md'
|
|
24
30
|
- '{process_slug}/interviews/{session_id}/raw-insights.md'
|
|
31
|
+
references:
|
|
32
|
+
skill_prompt: .claude/skills/process-interview/SKILL.md
|
|
33
|
+
agent_guide: .claude/agents/pm.md
|
|
25
34
|
context:
|
|
26
35
|
goal: 将音频访谈录音转写为结构化文本,生成带时间戳的逐字稿、摘要和初始洞察,为后续需求澄清阶段提供输入。
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
- 禁止修改 {process_slug}/recordings/ 中的原始文件。
|
|
30
|
-
|
|
31
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
32
|
-
|
|
33
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
36
|
+
execution: 调用 process-interview skill 处理 {process_slug}/recordings/ 下的原始音频,生成带时间戳的逐字稿、摘要与初始洞察,写入 {process_slug}/interviews/{session_id}/ 目录作为 clarify 阶段的输入。
|
|
34
37
|
gates:
|
|
35
38
|
entry:
|
|
36
39
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-analyze-frameworks
|
|
3
3
|
name: 分析框架应用
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 选择并应用 SWOT、波特五力、商业模式画布、PEST 等分析框架,规划可视化方案。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-competitive-analysis
|
|
8
8
|
next_stage: lc-collect-intelligence
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-analyze-frameworks
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/analysis-frameworks.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-analyze-frameworks/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 选择并应用 SWOT、波特五力、商业模式画布、PEST 等分析框架,规划可视化方案。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-analyze-frameworks skill 基于已收集的研究材料选择并应用合适的分析框架(SWOT、波特五力、商业模式画布、PEST 等),输出 analysis-frameworks.md 并规划 ECharts/HTML 可视化方案。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-collect-intelligence
|
|
3
3
|
name: 信息收集
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 基于权威报告、官方动态、开放 API 等渠道体系化收集研究证据并标注来源。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-analyze-frameworks
|
|
8
8
|
next_stage: lc-storytelling
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-collect-intelligence
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/collected-intelligence.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-collect-intelligence/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 基于权威报告、官方动态、开放 API 等渠道体系化收集研究证据并标注来源。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-collect-intelligence skill 基于权威报告、官方动态、开放 API、官网等渠道体系化收集研究证据,标注来源并写入 collected-intelligence.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-competitive-analysis
|
|
3
3
|
name: 竞品分析
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 分析竞争格局、竞争对手定位与差异化机会。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-product-research
|
|
8
8
|
next_stage: lc-analyze-frameworks
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-competitive-analysis
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/competitive.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-competitive-analysis/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 分析竞争格局、竞争对手定位与差异化机会。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-competitive-analysis skill 分析竞争格局、竞争对手定位、能力与差异化机会,输出 competitive.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-first-principles
|
|
3
3
|
name: 第一性原理探索
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 挑战现有假设,挖掘底层真相,为研究建立第一性原理基础。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-research-scope
|
|
8
8
|
next_stage: lc-stakeholder-research
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-first-principles
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-pm
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/first-principles.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-first-principles/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/pm.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 挑战现有假设,挖掘底层真相,为研究建立第一性原理基础。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-first-principles skill 帮助 PM 回归底层真相、挑战假设、识别第一性原理,输出 first-principles.md 作为研究的基础。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-market-research
|
|
3
3
|
name: 市场研究
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 分析市场规模、趋势、细分与动态,为产品定位提供事实基础。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-stakeholder-research
|
|
8
8
|
next_stage: lc-product-research
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-market-research
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/market.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-market-research/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 分析市场规模、趋势、细分与动态,为产品定位提供事实基础。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-market-research skill 分析市场规模、趋势、细分与动态,输出 market.md,为产品定位与机会判断提供事实基础。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-product-research
|
|
3
3
|
name: 产品研究
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 研究相关产品与解决方案的能力、价值主张、使用场景与优劣势。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-market-research
|
|
8
8
|
next_stage: lc-competitive-analysis
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-product-research
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/product-research.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-product-research/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 研究相关产品与解决方案的能力、价值主张、使用场景与优劣势。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-product-research skill 研究相关产品与解决方案的能力、价值主张、使用场景与优劣势,输出 product-research.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-research-report
|
|
3
3
|
name: 研究报告
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 整合全流程产物,输出可交付、可沉淀的最终研究报告。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-storytelling
|
|
8
8
|
next_stage: null
|
|
9
9
|
human_gate: true
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-research-report
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-pm
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/report.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-research-report/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/pm.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 整合全流程产物,输出可交付、可沉淀的最终研究报告。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-research-report skill 整合研究全流程产物(scope/first-principles/stakeholders/market/product-research/competitive/analysis-frameworks/collected-intelligence/narrative),输出可交付、可沉淀的 report.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-research-scope
|
|
3
3
|
name: 界定研究范围
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 与 PM 澄清研究目标、问题、范围与验收标准,输出已确认的研究 brief。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: workflow-router
|
|
8
8
|
next_stage: lc-first-principles
|
|
9
9
|
human_gate: true
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-research-scope
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-pm
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/scope.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-research-scope/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/pm.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 与 PM 澄清研究目标、问题、范围与验收标准,输出已确认的研究 brief。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-research-scope skill 与 PM 多轮澄清研究目标、问题、范围与验收标准,输出已确认的 scope.md,作为后续各研究阶段的单一事实来源。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-stakeholder-research
|
|
3
3
|
name: 相关者研究
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 识别并研究用户、买单方、建设方、维护方等关键相关者及其诉求。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-first-principles
|
|
8
8
|
next_stage: lc-market-research
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-stakeholder-research
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-researcher
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/stakeholders.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-stakeholder-research/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/researcher.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 识别并研究用户、买单方、建设方、维护方等关键相关者及其诉求。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-stakeholder-research skill 识别并研究用户、买单方、建设方、维护方等关键相关者,找到他们的发声渠道并分析其诉求,输出 stakeholders.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,12 +1,17 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: lc-storytelling
|
|
3
3
|
name: 体系化论述
|
|
4
|
-
description:
|
|
5
|
-
|
|
4
|
+
description: 基于金字塔原理将研究成果结构化为可说服决策者的故事体系。
|
|
5
|
+
workflows:
|
|
6
6
|
- pm-research
|
|
7
7
|
prerequisite_stage: lc-collect-intelligence
|
|
8
8
|
next_stage: lc-research-report
|
|
9
9
|
human_gate: false
|
|
10
|
+
execution_mode: solo
|
|
11
|
+
primary_action: lc-storytelling
|
|
12
|
+
variables:
|
|
13
|
+
- process_slug
|
|
14
|
+
- session_id
|
|
10
15
|
agent:
|
|
11
16
|
primary: lc-pm
|
|
12
17
|
reviewers:
|
|
@@ -19,15 +24,13 @@ skills:
|
|
|
19
24
|
artifacts:
|
|
20
25
|
required:
|
|
21
26
|
- '{process_slug}/research/{session_id}/narrative.md'
|
|
27
|
+
references:
|
|
28
|
+
skill_prompt: .claude/skills/lc-storytelling/SKILL.md
|
|
29
|
+
agent_guide: .claude/agents/pm.md
|
|
22
30
|
context:
|
|
23
31
|
goal: 基于金字塔原理将研究成果结构化为可说服决策者的故事体系。
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
- 本阶段产物写入 {process_slug}/research/{session_id}/。
|
|
27
|
-
|
|
28
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
29
|
-
|
|
30
|
-
- 详细执行步骤、模板与工具见对应 skill prompt 和 .claude/agents/<primary_agent>.md。'
|
|
32
|
+
execution: "调用 lc-storytelling skill 基于金字塔原理与科学论述逻辑,将研究成果结构化为可说服决策者的故事,输出 narrative.md。"
|
|
33
|
+
constraints: "- 本阶段产物写入 {process_slug}/research/{session_id}/。"
|
|
31
34
|
gates:
|
|
32
35
|
entry:
|
|
33
36
|
- check: previous_stage_completed
|
|
@@ -1,13 +1,31 @@
|
|
|
1
|
-
schema_version:
|
|
1
|
+
schema_version: 3.0.0
|
|
2
2
|
id: product-design-docs
|
|
3
3
|
name: Product Design Docs
|
|
4
|
-
description:
|
|
5
|
-
|
|
6
|
-
|
|
7
|
-
|
|
4
|
+
description: 基于已确认需求生成产品设计评审文档包(design-review、scenarios、feature-catalog、data-model、flows、feasibility、page-map、version-log、api-list 等 HTML),并完成 PM→UX 交接产物。
|
|
5
|
+
workflows:
|
|
6
|
+
- bug-fix
|
|
7
|
+
- design-spike
|
|
8
|
+
- existing-project-iteration
|
|
9
|
+
- interview-to-knowledge
|
|
10
|
+
- oss-first-design
|
|
8
11
|
prerequisite_stage: clarify
|
|
9
12
|
next_stage: product-prototype
|
|
10
13
|
human_gate: true
|
|
14
|
+
execution_mode: both
|
|
15
|
+
primary_action: draft-product-design
|
|
16
|
+
variables:
|
|
17
|
+
- process_slug
|
|
18
|
+
artifact_templates:
|
|
19
|
+
'{process_slug}/pages/docs/design-review.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
20
|
+
'{process_slug}/pages/docs/scenarios.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
21
|
+
'{process_slug}/pages/docs/feature-catalog.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
22
|
+
'{process_slug}/pages/docs/data-model.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
23
|
+
'{process_slug}/pages/docs/flows.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
24
|
+
'{process_slug}/pages/docs/feasibility.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
25
|
+
'{process_slug}/pages/docs/page-map.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
26
|
+
'{process_slug}/pages/docs/version-log.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
27
|
+
'{process_slug}/pages/docs/api-list.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
28
|
+
'{process_slug}/pages/docs/handoff-pm-to-ux-v*.html': .claude/templates/issue-package/page-doc-structured.html.tpl
|
|
11
29
|
agent:
|
|
12
30
|
primary: lc-pm
|
|
13
31
|
reviewers:
|
|
@@ -37,21 +55,15 @@ artifacts:
|
|
|
37
55
|
- '{process_slug}/pages/docs/handoff-pm-to-ux-v*.html'
|
|
38
56
|
- '{process_slug}/handoffs/pm-to-ux/pm-to-ux.handoff.yaml'
|
|
39
57
|
- '{process_slug}/handoffs/pm-to-ux/master-handoff-pm-to-ux-v*.md'
|
|
58
|
+
references:
|
|
59
|
+
skill_prompt: .claude/skills/draft-product-design/SKILL.md
|
|
60
|
+
agent_guide: .claude/agents/pm.md
|
|
40
61
|
context:
|
|
41
|
-
goal: 将已确认的需求转化为面向人类 PM
|
|
42
|
-
|
|
43
|
-
|
|
44
|
-
- 禁止修改 {process_slug}/recordings/ 中的原始文件。
|
|
45
|
-
|
|
46
|
-
- 本阶段设有 human_gate,必须获得人类 PM 显式确认后才能进入下一阶段。
|
|
47
|
-
|
|
48
|
-
- 本阶段 exit gate 要求 `handoffs/pm-to-ux/pm-to-ux.handoff.yaml` 存在。
|
|
49
|
-
|
|
50
|
-
- handoff_to 为 lc-designer;接收方 Agent 必须从 `handoffs/pm-to-ux/pm-to-ux.handoff.yaml` 开始读取,再按 `context_pack` 顺序读取 Tier-2 文档。
|
|
51
|
-
|
|
52
|
-
- 门控评估优先使用 AI 能力:读取产物并推理其完整性,而非仅依赖脚本化检查。
|
|
62
|
+
goal: 将已确认的需求转化为面向人类 PM 评审的简洁产品设计文档包(design-review/scenarios/feature-catalog/data-model/flows/feasibility/page-map/version-log/api-list 等 HTML),并完成 PM→UX 交接产物(agent-agent 契约),使人类 UX 能基于统一认知驱动 UX Agent。所有产物通过 {process_slug}/index.html 门户访问。
|
|
63
|
+
execution: 调用 draft-product-design skill 基于已确认的需求生成设计文档包(HTML 页面),并产出 PM→UX handoff 契约(pm-to-ux.handoff.yaml + master-handoff-pm-to-ux-v*.md),作为 UX 阶段接收方的入口。UI 规范、字段规范与可交互原型由下一阶段的 product-prototype 负责,不在本阶段产出。
|
|
64
|
+
constraints: '- 本阶段 exit gate 要求 `handoffs/pm-to-ux/pm-to-ux.handoff.yaml` 存在。
|
|
53
65
|
|
|
54
|
-
-
|
|
66
|
+
- handoff_to 为 lc-designer;接收方 Agent 必须从 `handoffs/pm-to-ux/pm-to-ux.handoff.yaml` 开始读取,再按 `context_pack` 顺序读取 Tier-2 文档。'
|
|
55
67
|
gates:
|
|
56
68
|
entry:
|
|
57
69
|
- check: previous_stage_completed
|