refs to click/type/drag/select
3. Re-snapshot if the page changed
## Skills
### /review-code - CodeReview After Changes
Reviewcode changes in the current branch compared to main. Provide structured feedback
conventions live in [CLAUDE.md](../CLAUDE.md). Both this file and CLAUDE.md are read by Copilot codereview (cap ~4,000 chars each). This file lists the rules worth named attention
verify no ESLint errors
### TDD Approach for Bug Fixes
When fixing bugs identified in codereviews or reported issues:
1. **Write failing test first**: Create a test that demonstrates
Copilot instructions for this repository
## High level guidance
* Review the `CONTRIBUTING.md` file for instructions to build and test the software.
* Run the `.github/Prime-ForCopilot.ps1` script (once) before running any `dotnet
affected LISA features
3. **Image Selection** - Choose appropriate Azure marketplace images
## Copilot Instructions for CodeReviews
**MANDATORY REQUIREMENT:** Every codereview must include a test suggestion using the Quick Reference
loops and inefficient patterns
- Prefer optimized patterns and consider memoization for expensive operations
---
## 📋 CodeReview Checklist
Before suggesting code, ensure:
- [ ] Only official APIs and standard library features are used
enum in `src/fabric_cicd/constants.py`. Always reference that file for the current list — do not hard-code item type strings without verifying them against the enum.
The publish/unpublish dependency order is defined
GitHub Copilot Instructions: Agent Skills Expert
Reviewcode changes as an **Agent Skills expert**, focusing on the [Agent Skills open format](https://agentskills.io/home) specification and best practices.
## Core Agent
Type vResult = this.FragmentFactory.CreateFragment ()] { ... } : ( alternatives ) ;` — this pattern initializes an AST fragment via FragmentFactory.
- Parser-generated code frequently uses `Match( , CodeGenerationSupporter. )` and `ThrowParseErrorException("SQLxxxx", ...)` for diagnostics.
- The codebase prefers using the factory
code before confirmation.
- If the user requests a UI redesign or says the UI is unsatisfactory, first update `output/*-uiux.md`, then redo the frontend, and rerun frontend runtime + UI review
version (e.g., `app-name@1.0.0`)
- List all Azure services and dependencies
- Include infrastructure as code (Bicep or Terraform) configuration
Example azure.yaml structure:
```yaml
name: my-ai-app
metadata:
template
These are
build artifacts; they must not be edited or committed.
## Review priorities
When reviewing pull requests, focus on:
- Correctness of the Bun bridge: HTTP/WebSocket message handling, socket
lifecycle, reconnection
account show --query name -o tsv`) and key decisions off the command's exit code.
- **Keep CLI install hints in the repo's standard call form**, e.g. `mcp_azure
Create checkpoint at release points
empirica checkpoint-create --session-id --message "Feature complete"
```
**CodeReview Patterns:**
1. PREFLIGHT before review - assess familiarity with codebase area
2. Log unknowns for areas