Sibling skills (local only)
Sibling CloudBase skills ship beside this skill. Use local relative paths such as ../auth-tool-cloudbase/SKILL.md.
If a referenced sibling skill file is missing from this environment, ask the user to install the full CloudBase plugin (or the missing skill). Do not HTTP-fetch remote skill or protocol markdown into the agent context.
Spec Workflow
Activation Contract
Use this first when
- The request is a new feature, multi-step product change, cross-module integration, or architecture/design task.
- Acceptance criteria are unclear and need to be made explicit before implementation.
- The work involves multiple files, user flows, database design, or UI design that needs staged confirmation.
Read before writing code if
- You are unsure whether the task should go straight to coding or should first go through requirements, design, and task planning.
- The request mentions a new page, a new system, a redesign, a workflow, or a multi-module refactor.
Then also read
- Frontend page or visual design work ->
../ui-design/SKILL.md
- Advanced data-model work ->
../data-model-creation/SKILL.md
Do NOT use for
- Small bug fixes with clear scope.
- One-file documentation updates.
- Straightforward config changes.
- Tiny refactors where the user already gave exact implementation instructions.
Common mistakes / gotchas
- Jumping into coding before acceptance criteria are explicit.
- Skipping user confirmation between requirements, design, and tasks.
- Writing vague tasks that do not map back to user-visible outcomes.
- Treating UI work as purely technical implementation without clarifying design intent.
Minimal checklist
- Decide whether the change really needs the full spec flow.
- If yes, stop and produce requirements first.
- If the change is small, low-risk, and acceptance is already clear, allow direct execution without forcing spec artifacts.
- Use EARS-style acceptance criteria.
- Get confirmation before moving to the next phase.
When to use this skill
Use this workflow for structured development when you need to:
- Define or refine a new feature
- Design complex architecture
- Coordinate changes across modules
- Plan database or UI-heavy work
- Improve requirement quality and acceptance boundaries
Decision rule
Use the full workflow when
- The task is medium or large
- The impact spans multiple modules
- Acceptance boundaries are fuzzy
- The user wants disciplined planning before implementation
Skip the full workflow when
- The task is small, low-risk, and already precise
- Goal, scope, and acceptance are already clear enough to execute directly
- The user explicitly wants a direct code change with no planning phase
Core workflow
Phase 1: Requirements
Create specs/<spec_name>/requirements.md.
What to do:
- Restate the problem and scope
- Write user stories
- Write acceptance criteria in EARS style
- Clarify business rules, constraints, and non-goals
EARS pattern:
While <optional precondition>, when <optional trigger>, the <system name> shall <system response>
Example:
When the user submits the form, the booking system shall validate required fields before creating the record.
Phase 2: Design
Create specs/<spec_name>/design.md.
What to do:
- Describe architecture and module boundaries
- Explain technology choices and trade-offs
- Define data model, API, security, and testing strategy as needed
- Use Mermaid only when a diagram materially improves clarity
Phase 3: Tasks
Create specs/<spec_name>/tasks.md.
What to do:
- Break the design into executable tasks
- Keep tasks specific and reviewable
- Link each task back to the relevant requirement
- Update task status as work progresses
Task format:
# Implementation Plan
- [ ] 1. Task title
- Specific work item
- Another concrete step
- _Requirement: 1
Phase 4: Execution
Only start implementation after the user confirms the task plan.
During execution:
- Keep task status current
- Finish one meaningful unit at a time
- Preserve traceability from change -> task -> requirement
Working rules for the agent
- Ask follow-up questions when the request is underspecified; do not guess core product behavior.
- Require confirmation between requirements, design, and task breakdown.
- Pull in
ui-design early when the change includes end-user pages or visual decisions.
- Keep documents concise but testable.
- Prefer user-visible outcomes over implementation-detail task names.
Output expectations
requirements.md -> problem, scope, user stories, EARS acceptance criteria
design.md -> architecture, technical approach, data/API/security/test notes
tasks.md -> actionable implementation checklist tied to requirements
1---2name: spec-workflow3description: Use when medium-to-large changes need explicit requirements, technical design, and task planning before implementation, especially for multi-module work, unclear acceptance criteria, or architecture-heavy requests.4---5
6## Sibling skills (local only)
7
8Sibling CloudBase skills ship beside this skill. Use local relative paths such as `../auth-tool-cloudbase/SKILL.md`.
9
10If a referenced sibling skill file is missing from this environment, ask the user to install the full CloudBase plugin (or the missing skill). Do **not** HTTP-fetch remote skill or protocol markdown into the agent context.
11
12# Spec Workflow
13
14## Activation Contract
15
16### Use this first when
17
18- The request is a new feature, multi-step product change, cross-module integration, or architecture/design task.
19- Acceptance criteria are unclear and need to be made explicit before implementation.
20- The work involves multiple files, user flows, database design, or UI design that needs staged confirmation.
21
22### Read before writing code if
23
24- You are unsure whether the task should go straight to coding or should first go through requirements, design, and task planning.
25- The request mentions a new page, a new system, a redesign, a workflow, or a multi-module refactor.
26
27### Then also read
28
29- Frontend page or visual design work -> `../ui-design/SKILL.md`
30- Advanced data-model work -> `../data-model-creation/SKILL.md`
31
32### Do NOT use for
33
34- Small bug fixes with clear scope.
35- One-file documentation updates.
36- Straightforward config changes.
37- Tiny refactors where the user already gave exact implementation instructions.
38
39### Common mistakes / gotchas
40
41- Jumping into coding before acceptance criteria are explicit.
42- Skipping user confirmation between requirements, design, and tasks.
43- Writing vague tasks that do not map back to user-visible outcomes.
44- Treating UI work as purely technical implementation without clarifying design intent.
45
46### Minimal checklist
47
48- Decide whether the change really needs the full spec flow.
49- If yes, stop and produce requirements first.
50- If the change is small, low-risk, and acceptance is already clear, allow direct execution without forcing spec artifacts.
51- Use EARS-style acceptance criteria.
52- Get confirmation before moving to the next phase.
53
54## When to use this skill
55
56Use this workflow for structured development when you need to:
57
58- Define or refine a new feature
59- Design complex architecture
60- Coordinate changes across modules
61- Plan database or UI-heavy work
62- Improve requirement quality and acceptance boundaries
63
64## Decision rule
65
66### Use the full workflow when
67
68- The task is medium or large
69- The impact spans multiple modules
70- Acceptance boundaries are fuzzy
71- The user wants disciplined planning before implementation
72
73### Skip the full workflow when
74
75- The task is small, low-risk, and already precise
76- Goal, scope, and acceptance are already clear enough to execute directly
77- The user explicitly wants a direct code change with no planning phase
78
79## Core workflow
80
81### Phase 1: Requirements
82
83Create `specs/<spec_name>/requirements.md`.
84
85What to do:
86
87- Restate the problem and scope
88- Write user stories
89- Write acceptance criteria in EARS style
90- Clarify business rules, constraints, and non-goals
91
92EARS pattern:
93
94```text
95While <optional precondition>, when <optional trigger>, the <system name> shall <system response>
96```
97
98Example:
99
100```text
101When the user submits the form, the booking system shall validate required fields before creating the record.
102```
103
104### Phase 2: Design
105
106Create `specs/<spec_name>/design.md`.
107
108What to do:
109
110- Describe architecture and module boundaries
111- Explain technology choices and trade-offs
112- Define data model, API, security, and testing strategy as needed
113- Use Mermaid only when a diagram materially improves clarity
114
115### Phase 3: Tasks
116
117Create `specs/<spec_name>/tasks.md`.
118
119What to do:
120
121- Break the design into executable tasks
122- Keep tasks specific and reviewable
123- Link each task back to the relevant requirement
124- Update task status as work progresses
125
126Task format:
127
128```markdown
129# Implementation Plan
130
131- [ ] 1. Task title
132 - Specific work item
133 - Another concrete step
134 - _Requirement: 1
135```
136
137### Phase 4: Execution
138
139Only start implementation after the user confirms the task plan.
140
141During execution:
142
143- Keep task status current
144- Finish one meaningful unit at a time
145- Preserve traceability from change -> task -> requirement
146
147## Working rules for the agent
148
1491. Ask follow-up questions when the request is underspecified; do not guess core product behavior.
1502. Require confirmation between requirements, design, and task breakdown.
1513. Pull in `ui-design` early when the change includes end-user pages or visual decisions.
1524. Keep documents concise but testable.
1535. Prefer user-visible outcomes over implementation-detail task names.
154
155## Output expectations
156
157- `requirements.md` -> problem, scope, user stories, EARS acceptance criteria
158- `design.md` -> architecture, technical approach, data/API/security/test notes
159- `tasks.md` -> actionable implementation checklist tied to requirements