create-workflow-command

Name: create-workflow-command
Author: NeoLabHQ

817installs132forks

How to install create-workflow-command?

neolabhq/context-engineering-kit/create-workflow-command

$npx -y skills add NeoLabHQ/context-engineering-kit --skill create-workflow-command

Installs into the current project.

Summary

Create a workflow command that orchestrates multi-step execution through sub-agents with file-based task prompts

Files · 1

View on GitHub

SKILL.md

1	# Create Workflow Command
2
3	Create a command that orchestrates multi-step workflows by dispatching sub-agents with task-specific instructions stored in separate files.
4
5	## User Input
6
7	```text
8	Workflow Name: $1
9	Description: $2
10	```
11
12	## Architecture Overview
13
14	Workflow commands solve the context bloat problem: instead of embedding detailed step instructions in the main command (polluting orchestrator context), store them in separate task files that sub-agents read on-demand.
15
16	```
17	plugins/<plugin-name>/
18	├── commands/
19	│ └── <workflow>.md # Lean orchestrator (~50-100 tokens per step)
20	├── agents/ # Optional: reusable executor agents
21	│ └── step-executor.md # Custom agent with specific tools/behavior
22	└── tasks/ # All task instructions directly here
23	├── step-1-<name>.md # Full instructions (~500+ tokens each)
24	├── step-2-<name>.md
25	├── step-3-<name>.md
26	└── common-context.md # Shared context across workflows
27	```
28
29	## Key Principles
30
31	### 1. Context Isolation
32
33	Each sub-agent gets its own isolated context window. The main orchestrator stays lean while sub-agents load detailed instructions from files.
34
35	\| Component \| Context Cost \| Purpose \|
36	\|-----------\|--------------\|---------\|
37	\| Orchestrator command \| ~50-100 tokens/step \| Dispatch and coordinate \|
38	\| Task file \| ~500+ tokens \| Detailed step instructions \|
39	\| Sub-agent base \| ~294 tokens \| System prompt overhead \|
40
41	### 2. Sub-Agent Capabilities
42
43	Sub-agents spawned via Task tool:
44
45	\| Capability \| Available \| Notes \|
46	\|------------\|-----------\|-------\|
47	\| Read tool \| ✅ Yes \| Can read any file \|
48	\| Write tool \| ✅ Yes \| If not restricted \|
49	\| Grep/Glob \| ✅ Yes \| For code search \|
50	\| Skills loading \| ❌ No \| Skills don't auto-load in sub-agents \|
51	\| Spawn sub-agents \| ❌ No \| Cannot nest Task tool \|
52	\| Resume context \| ✅ Yes \| Via `resume` parameter \|
53
54	### 3. File Reference Pattern
55
56	Use `${CLAUDE_PLUGIN_ROOT}` for portable paths within plugin:
57
58	```markdown
59	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-1-workflow-name.md and execute.
60	```
61
62	Sub-agent will use Read tool to fetch the file content.
63
64	## Implementation Process
65
66	### Step 1: Gather Requirements
67
68	Ask user (if not provided):
69
70	1. Workflow name: kebab-case identifier (e.g., `feature-implementation`)
71	2. Description: What the workflow accomplishes
72	3. Steps: List of discrete steps with:
73	- Step name
74	- Step goal
75	- Required tools
76	- Expected output
77	4. Execution mode: Sequential or parallel steps
78	5. Agent type: `general-purpose` or custom agent
79
80	### Step 2: Create Directory Structure
81
82	```bash
83	# Create tasks directory (if it doesn't exist)
84	mkdir -p ${CLAUDE_PLUGIN_ROOT}/tasks
85
86	# Optional: Create agents directory (if using custom agents)
87	mkdir -p ${CLAUDE_PLUGIN_ROOT}/agents
88	```
89
90	Note: All task files (both workflow-specific steps and shared context) are placed directly in `tasks/` without subdirectories.
91
92	### Step 3: Create Task Files
93
94	For each step, create a task file with this structure:
95
96	```markdown
97	# Step N: <Step Name>
98
99	## Context
100	You are executing step N of the <workflow-name> workflow.
101
102	## Goal
103	<Clear, specific goal for this step>
104
105	## Input
106	<What this step receives from previous steps or user>
107
108	## Instructions
109	1. <Specific action>
110	2. <Specific action>
111	3. <Specific action>
112
113	## Constraints
114	- <Limitation or boundary>
115	- <What NOT to do>
116
117	## Expected Output
118	<What to return to orchestrator>
119
120	## Success Criteria
121	- [ ] <Measurable outcome>
122	- [ ] <Measurable outcome>
123	```
124
125	### Step 4: Create Orchestrator Command
126
127	Create the main command file with this pattern:
128
129	```markdown
130	---
131	description: <Workflow description>
132	argument-hint: <Required arguments>
133	allowed-tools: Task, Read
134	model: sonnet
135	---
136
137	# <Workflow Name>
138
139	## User Input
140
141	\`\`\`text
142	$ARGUMENTS
143	\`\`\`
144
145	## Workflow Execution
146
147	### Step 1: <Step Name>
148
149	Launch general-purpose agent:
150	- Description: "<3-5 word summary>"
151	- Prompt:
152	\`\`\`
153	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-1-<workflow>-<name>.md and execute.
154
155	Context:
156	- TARGET: $1
157	- MODE: $2
158	\`\`\`
159
160	Capture: <What to extract from result>
161
162	### Step 2: <Step Name>
163
164	Launch general-purpose agent:
165	- Description: "<3-5 word summary>"
166	- Prompt:
167	\`\`\`
168	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-2-<workflow>-<name>.md and execute.
169
170	Context from Step 1:
171	- <Key data from previous step>
172	\`\`\`
173
174	### Step 3: <Step Name>
175
176	[Continue pattern...]
177
178	## Completion
179
180	Summarize workflow results:
181	1. <What was accomplished>
182	2. <Key outputs>
183	3. <Next steps if any>
184	```
185
186	#### Frontmatter Options
187
188	\| Field \| Purpose \| Default \|
189	\|-------\|---------\|---------\|
190	\| `description` \| Brief description of workflow purpose \| Required \|
191	\| `argument-hint` \| Expected arguments descrip

People who liked this also viewed

6 picks

Type

lark-drive飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sSkillsJul 2026382k16k
lark-sharedUse for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handlingSkillsJul 2026382k16k
lark-whiteboard飞书画板：查询和编辑飞书云文档中的画板。支持导出画板为预览图片、导出原始节点结构、使用多种格式更新画板内容。当用户需要查看画板内容、导出画板图片、编辑画板时使用此 skill。不负责：飞书云文档内容编辑（lark-doc）、文档内嵌电子表格/Base（lark-sheets / lark-base）。SkillsJul 2026380k16k
lark-task飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查看任务列表、跟踪任务进度、管理项目清单或给他人分配任务、为任务上传附件文件、注册注销任务智能体、更新智能体主页数据、写入任务记录时使用。SkillsJul 2026380k16k
lark-attendance飞书考勤打卡：查询自己的考勤打卡记录SkillsJul 2026330k16k
lark-apps妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用角色与成员管理、自动化触发器（定时/记录变更/Webhook/飞书审批）。当用户要SkillsJul 2026227k16k

create-workflow-command

817installs132forks

View on GitHub

How to install create-workflow-command?

neolabhq/context-engineering-kit/create-workflow-command

$npx -y skills add NeoLabHQ/context-engineering-kit --skill create-workflow-command

Installs into the current project.

Summary

Create a workflow command that orchestrates multi-step execution through sub-agents with file-based task prompts

Files · 1

View on GitHub

SKILL.md

1	# Create Workflow Command
2
3	Create a command that orchestrates multi-step workflows by dispatching sub-agents with task-specific instructions stored in separate files.
4
5	## User Input
6
7	```text
8	Workflow Name: $1
9	Description: $2
10	```
11
12	## Architecture Overview
13
14	Workflow commands solve the context bloat problem: instead of embedding detailed step instructions in the main command (polluting orchestrator context), store them in separate task files that sub-agents read on-demand.
15
16	```
17	plugins/<plugin-name>/
18	├── commands/
19	│ └── <workflow>.md # Lean orchestrator (~50-100 tokens per step)
20	├── agents/ # Optional: reusable executor agents
21	│ └── step-executor.md # Custom agent with specific tools/behavior
22	└── tasks/ # All task instructions directly here
23	├── step-1-<name>.md # Full instructions (~500+ tokens each)
24	├── step-2-<name>.md
25	├── step-3-<name>.md
26	└── common-context.md # Shared context across workflows
27	```
28
29	## Key Principles
30
31	### 1. Context Isolation
32
33	Each sub-agent gets its own isolated context window. The main orchestrator stays lean while sub-agents load detailed instructions from files.
34
35	\| Component \| Context Cost \| Purpose \|
36	\|-----------\|--------------\|---------\|
37	\| Orchestrator command \| ~50-100 tokens/step \| Dispatch and coordinate \|
38	\| Task file \| ~500+ tokens \| Detailed step instructions \|
39	\| Sub-agent base \| ~294 tokens \| System prompt overhead \|
40
41	### 2. Sub-Agent Capabilities
42
43	Sub-agents spawned via Task tool:
44
45	\| Capability \| Available \| Notes \|
46	\|------------\|-----------\|-------\|
47	\| Read tool \| ✅ Yes \| Can read any file \|
48	\| Write tool \| ✅ Yes \| If not restricted \|
49	\| Grep/Glob \| ✅ Yes \| For code search \|
50	\| Skills loading \| ❌ No \| Skills don't auto-load in sub-agents \|
51	\| Spawn sub-agents \| ❌ No \| Cannot nest Task tool \|
52	\| Resume context \| ✅ Yes \| Via `resume` parameter \|
53
54	### 3. File Reference Pattern
55
56	Use `${CLAUDE_PLUGIN_ROOT}` for portable paths within plugin:
57
58	```markdown
59	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-1-workflow-name.md and execute.
60	```
61
62	Sub-agent will use Read tool to fetch the file content.
63
64	## Implementation Process
65
66	### Step 1: Gather Requirements
67
68	Ask user (if not provided):
69
70	1. Workflow name: kebab-case identifier (e.g., `feature-implementation`)
71	2. Description: What the workflow accomplishes
72	3. Steps: List of discrete steps with:
73	- Step name
74	- Step goal
75	- Required tools
76	- Expected output
77	4. Execution mode: Sequential or parallel steps
78	5. Agent type: `general-purpose` or custom agent
79
80	### Step 2: Create Directory Structure
81
82	```bash
83	# Create tasks directory (if it doesn't exist)
84	mkdir -p ${CLAUDE_PLUGIN_ROOT}/tasks
85
86	# Optional: Create agents directory (if using custom agents)
87	mkdir -p ${CLAUDE_PLUGIN_ROOT}/agents
88	```
89
90	Note: All task files (both workflow-specific steps and shared context) are placed directly in `tasks/` without subdirectories.
91
92	### Step 3: Create Task Files
93
94	For each step, create a task file with this structure:
95
96	```markdown
97	# Step N: <Step Name>
98
99	## Context
100	You are executing step N of the <workflow-name> workflow.
101
102	## Goal
103	<Clear, specific goal for this step>
104
105	## Input
106	<What this step receives from previous steps or user>
107
108	## Instructions
109	1. <Specific action>
110	2. <Specific action>
111	3. <Specific action>
112
113	## Constraints
114	- <Limitation or boundary>
115	- <What NOT to do>
116
117	## Expected Output
118	<What to return to orchestrator>
119
120	## Success Criteria
121	- [ ] <Measurable outcome>
122	- [ ] <Measurable outcome>
123	```
124
125	### Step 4: Create Orchestrator Command
126
127	Create the main command file with this pattern:
128
129	```markdown
130	---
131	description: <Workflow description>
132	argument-hint: <Required arguments>
133	allowed-tools: Task, Read
134	model: sonnet
135	---
136
137	# <Workflow Name>
138
139	## User Input
140
141	\`\`\`text
142	$ARGUMENTS
143	\`\`\`
144
145	## Workflow Execution
146
147	### Step 1: <Step Name>
148
149	Launch general-purpose agent:
150	- Description: "<3-5 word summary>"
151	- Prompt:
152	\`\`\`
153	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-1-<workflow>-<name>.md and execute.
154
155	Context:
156	- TARGET: $1
157	- MODE: $2
158	\`\`\`
159
160	Capture: <What to extract from result>
161
162	### Step 2: <Step Name>
163
164	Launch general-purpose agent:
165	- Description: "<3-5 word summary>"
166	- Prompt:
167	\`\`\`
168	Read ${CLAUDE_PLUGIN_ROOT}/tasks/step-2-<workflow>-<name>.md and execute.
169
170	Context from Step 1:
171	- <Key data from previous step>
172	\`\`\`
173
174	### Step 3: <Step Name>
175
176	[Continue pattern...]
177
178	## Completion
179
180	Summarize workflow results:
181	1. <What was accomplished>
182	2. <Key outputs>
183	3. <Next steps if any>
184	```
185
186	#### Frontmatter Options
187
188	\| Field \| Purpose \| Default \|
189	\|-------\|---------\|---------\|
190	\| `description` \| Brief description of workflow purpose \| Required \|
191	\| `argument-hint` \| Expected arguments descrip

People who liked this also viewed

6 picks

Type