60
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill cin12211/orca-q --skill code-review- SKILL.md12.5 KB
Overview
This skill performs deep, context-aware code reviews focused on architecture, code quality, security, performance, testing, and documentation. It delivers prioritized, actionable feedback with root-cause analysis and working code suggestions tailored to a Vue-based database editor project. Use it after significant changes to ensure maintainability and safety before merging.
How this skill works
The reviewer first gathers repository context (directory layout, conventions, recent commits, config files) and maps domain concepts from identifiers and modules. It runs pattern detection across files to locate anti-patterns, duplication, hot paths, and security-sensitive areas. For every finding it provides three levels: what the issue is, why it occurs, and a concrete fix or code example, plus impact-based prioritization.
When to use it
- Before merging major feature branches (UI, DB model, or API changes)
- After refactors that touch controllers, services, or shared utilities
- When adding or upgrading dependencies, especially security-sensitive ones
- Before shipping performance-sensitive features (queries, caching, syncing)
- When test coverage or API contracts change
Best practices
- Provide repository context: reference relevant docs and recent commits before review
- Classify findings by impact (critical/high/medium/low) and focus first on security and data integrity
- Always include root cause analysis and at least one working code example or snippet
- Check cross-file relationships: imports, tests, interfaces, and docs when recommending changes
- Prioritize consolidation of duplicated logic into shared modules (validators, DB helpers)
- Recommend tests that assert behavior and edge cases, not only achieve coverage
Example use cases
- Review a new query layer that adds PostgreSQL stored-procedure calls to ensure separation of concerns and SQL safety
- Audit dependency upgrades (Electron, Nuxt, Vue) for breaking changes and supply-chain risks
- Analyze UI components for memory leaks and improper event cleanup in Electron windows
- Assess test suites for meaningful assertions around DB migrations and backup logic
- Evaluate API surface changes for breaking clients and missing documentation
FAQ
Each issue includes a concise explanation, root cause, and at least one concrete code example or pattern to apply in the codebase.
Will the review check for security vulnerabilities in dependencies?
Yes. The review highlights risky dependency upgrades, known supply-chain issues, and insecure usage patterns in auth, tokens, and DB access.
Can I run multiple focused reviews in parallel?
Yes. The skill supports running separate reviews for each of the six focus areas to get targeted feedback quickly.