en un clic
phpstan-rules
phpstan-rules contient 20 skills collectées depuis pekral, avec une couverture métier par dépôt et des pages de détail sur le site.
Skills dans ce dépôt
Use when analyzing problems from issue trackers. Downloads and reviews attachments, provides technical analysis and solutions, and creates human-readable explanations for both technical and non-technical audiences.
Use when refactoring PHP classes following Laravel best practices and SOLID principles. Ensures code quality, maintains functionality, improves testability, and achieves 100% code coverage. Focuses on single responsibility, DRY principle, and clean code structure.
Use when performing code review for GitHub pull requests. Analyzes code changes, identifies critical and moderate issues, runs tests, and posts review comments. Reviews code quality, security, and adherence to project standards.
Use when performing code review for JIRA issues. Analyzes pull requests, identifies critical and moderate issues, runs tests, and posts review comments to GitHub PRs. Reviews code quality, security, and adherence to project standards.
Senior PHP code reviewer. Use when reviewing pull requests, examining code changes vs master branch, or when the user asks for a code review. Read-only review — never modifies code.
Use when the user ran composer update and wants to check for package conflicts and get a summary of changelogs from updated packages.
Create an issue in a generic issue tracker from the provided task description. Use when the user asks to open an issue. Preserve the original task content exactly and only improve formatting for readability.
Reads your pull request code review, verifies that all recommended test coverage is implemented in the codebase, and adds missing tests using the create-test skill. Use when a PR review already exists and missing tests must be completed with 100% coverage for current changes.
Use when creating tests following project conventions and patterns. Ensures deterministic tests, 100% code coverage for changes, uses data providers where appropriate, and mocks only external services or exception scenarios.
Use when analyzing Laravel Telescope requests from URL and DB. Loads Telescope entries, matches the same request in database tables, and proposes practical optimizations.
Use when merging PRs that are ready for deployment, one by one.
Use when you need to analyze and solve MySQL performance or query design problems directly from a real codebase or database environment. Inspects SQL queries, table structure, index usage, and uses terminal tools (mysql, EXPLAIN) when possible. For pragmatic diagnosis of slow queries, joins, indexes, and filtering in existing applications.
Use when processing pull request code review feedback. Finds the latest PR for a task, resolves review comments, updates review status, and triggers the next review cycle.
Use when refactoring any entry point logic from a controller/job/command/listener into an Action class. Enforces Action pattern rules from project.mdc and related architecture rules.
Use when rewriting existing tests to PEST syntax. Follows project conventions, ensures DRY principles, uses data providers, maintains 100% coverage, and verifies test functionality.
Use when performing comprehensive security review for Laravel/PHP projects. Follows OWASP Top 10 and incident-response hardening checks (packages, malware/webshells, configuration, uploads, credentials), and provides structured reports with severity levels.
Use when you want one radically useful, high-impact project addition proposal.
Use when implementing any feature or bugfix using TDD methodology. Enforces red-green-refactor cycle, ensures failing test exists before any production code, prevents test-after anti-patterns, and maintains 100% coverage for changes.
Use when testing the current pull request. Find the 'Testing Recommendations' section in the PR conversation and test the application like a senior web application tester. Follow the described scenarios, use tools when needed, and produce a human-readable report without technical notes.
Use when the agent must follow a strict problem-solving loop: understand, propose, implement, verify.