You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: .github/copilot-instructions.md
+5-1Lines changed: 5 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,4 +16,8 @@
16
16
## Prisma and Database Guidelines
17
17
- If you need to run `npx prisma generate`, run `npm run db:generate` from `apps/backend-services` - it's a special script that writes models into apps/temporal/src and apps/backend-services/src. Don't forget to run migrations as normal if necessary.
18
18
- Tables should be designed with a `created_at` timestamp (default to now) and an `updated_at` timestamp (auto-updated on change) for auditing purposes.
19
-
- Table names should be singular (e.g., `User`, not `Users`) to align with Prisma conventions.
19
+
- Table names should be singular (e.g., `User`, not `Users`) to align with Prisma conventions.
20
+
21
+
## Requirements and User Stories
22
+
- When finished implementing a user story, check it off in the related user stories file in `feature-docs/002-group-management/user_stories/README.md` and update the acceptance checklist.
23
+
- If you find any gaps in the requirements or user stories, document them in the summary notes after implementation and ask for clarification before proceeding with any assumptions.
description: "Requirements Refiner: Iteratively questions the user to clarify vague requirements and outputs a single consolidated requirements document in feature-docs."
4
+
---
5
+
**Role**: You are an expert Technical Product Manager and Business Analyst. Your goal is to take initial, potentially vague business requirements and refine them into a clear, comprehensive specification by asking clarifying questions.
6
+
7
+
## Your Workflow
8
+
9
+
1.**Analyze**: Read the input text or file. Compare it against the standards defined in `.claude/skills/requirements-refiner/elicitation_standards.md`.
10
+
2.**Iterative Elicitation**:
11
+
- Identify gaps, ambiguities, or assumptions.
12
+
- Ask a set of numbered clarifying questions.
13
+
- Wait for the user's response.
14
+
- Repeat this step until the requirements are clear and complete.
15
+
3.**Consolidate**:
16
+
- Once the requirements are fully understood, create a single consolidated document.
17
+
- This document should include the initial requirements plus all details gathered during the Q&A process.
18
+
- Ensure the output is structured and ready for a User Story writer to consume.
19
+
4.**Save to feature-docs**:
20
+
- Examine the `feature-docs/` directory to find the highest numbered folder (e.g., `001-graph-workflows`).
21
+
- Determine the next increment number (e.g., if `001` exists, use `002`).
22
+
- Generate a short feature name/slug (kebab-case, e.g., "benchmarking-system").
23
+
- Create a new folder: `feature-docs/{NNN}-{feature-slug}/` (e.g., `feature-docs/002-benchmarking-system/`).
24
+
- Save the consolidated requirements as `REQUIREMENTS.md` in that folder.
25
+
26
+
## Key Behaviors
27
+
-**Iterative Approach**: Do not rush to the final output. Prioritize clarity over speed.
28
+
-**Probe Deeply**: Ask about edge cases, error states, and user roles.
29
+
-**Auto-increment Folders**: Always check existing feature-docs folders to determine the next number.
30
+
-**Output Format**: The final output must be saved as `feature-docs/{NNN}-{feature-slug}/REQUIREMENTS.md`.
description: Convert a refined requirements document into one user story file per story in a user_stories subfolder using the strict template, plus a README.md with phase breakdown.
4
+
argument-hint: refined-requirements-file
5
+
context: fork
6
+
agent: user-story-writer
7
+
disable-model-invocation: true
8
+
---
9
+
Generate user stories from: $ARGUMENTS
10
+
11
+
Requirements:
12
+
- Decompose into atomic, independent user stories.
13
+
- Write each story using the strict template you have access to.
14
+
- Create user_stories subfolder in the same directory as the requirements file.
15
+
- Use filename format `US-{number}-{short-description}.md`.
16
+
- Generate a README.md in the user_stories folder with all stories organized by phases with checkboxes.
0 commit comments