Review code changes against project conventions and architectural standards.
npx skills add https://github.com/highflame-ai/ai-factory --skill reflect-highflame-ai
Please help me install this Agent Skill. Skill: reflect Source: https://github.com/highflame-ai/ai-factory/tree/main/skills/reflect Command: npx skills add https://github.com/highflame-ai/ai-factory --skill reflect-highflame-ai
This Skill addresses the common issue of shipping code that lacks sufficient self-scrutiny, helping developers catch architectural flaws, convention violations, and testing gaps before formal review.
Run the reflect skill to perform a self-review of the changes on the current branch against the main branch.
High-intent search queries and answers about installing and using this skill.
To perform an automated code review, you can run the reflect skill against your current branch to systematically check for architectural, stylistic, and functional issues before formal submission.
Automated self-review is a systematic assessment of code changes against a canonical checklist to identify architecture violations, convention gaps, and testing issues prior to human review.
You can scope analysis to specific branches or requirements, utilizing a dedicated reflector agent to validate your code against project-wide conventions and architectural standards.
Yes, context-aware analysis integrates your project-specific conventions and architecture documentation to ensure your code changes comply with established team standards.
No, automated self-review acts as a pre-check to catch architectural flaws and testing gaps, ensuring your code is ready and reducing friction during the formal human review process.
Automated self-assessment scopes analysis to specific branches and validates against a canonical checklist, but relies on existing project conventions and cannot replace subjective human judgment in formal reviews.
AribSudiaGenerate JSDoc, Markdown API references, and architecture guides from source code.
silva2kandAnalyze local code changes or GitHub pull requests for issues and improvements.
Chang-Shih-YungAggregate type checker, linter, and test results into a 0-10 composite score.
ReidondReviews completed dev tasks for code quality issues and documentation gaps.
