AI Blocks › All skills by job

Test and fix code

142 free AI skills and workflows for this job.

• Accessibility checker: Point it at a page or code. Get ranked WCAG 2.2 problems, with fixes if you ask.
• Agent run evidence auditor: Give it logs from an AI agent run. Get each success claim checked against real proof, with gaps marked.
• Agent system tester: Describe your AI agent. Get a way to measure how well it works and compare changes.
• AI agent tester: Describe your AI agent. Get a test set, scoring rules and a release decision memo you can trust.
• AI answer judge designer: Say what good looks like. Get a scoring guide an AI can use to grade answers fairly, with bias checks.
• AI answer testing planner: Describe your AI feature. Get a test plan with scores, human review and A/B checks to catch weak answers.
• AI-written code reviewer: Give it code an AI wrote fast. Get the real risks that will hurt later, with the smallest fix for each.
• Animation reviewer: Give it a code change with motion. Get a table of fixes and a clear approve or block verdict.
• API contract tester: Give it an API and its users. Get tests that catch breaking changes between the two before they ship.
• API design reviewer: Give it your API spec. Get a scorecard, flagged breaking changes and fixes for naming and errors.
• API test pipeline writer: Give it your Postman collection. Get a ready CI pipeline file that runs your API tests on every push.
• API test suite builder: Point it at your endpoints. Get integration and contract tests for each route, with fixtures.
• App test coverage review: Point it at your app. Get every page, part and endpoint listed with what is tested and what is not.
• Browser automation scripter: Say what to do in a web browser. Get Playwright code written and run for it, with results and screenshots.
• Browser page debugger: Point it at a page that misbehaves. Get the real errors, network calls and slow spots found in a live browser.
• Browser task runner: Describe a browser task. Get pages visited, forms filled, and screenshots or data saved.
• Browser test migrator: Point it at Cypress or Selenium tests. Get them moved to Playwright file by file.
• Browser test reviewer: Point it at your Playwright tests. Get each one checked against 20 bad habits with fixes.
• Browser test setup builder: Point it at your project. Get Playwright installed with a config, a sample test and a CI file.
• Browser test writer: Describe a user story or page. Get Playwright tests written for it that you can run.
• BrowserStack test runner: Give it your Playwright tests. Get them run on many browsers and devices, with a pass and fail report.
• Bug and security code reviewer: Give it a file or pull request. Get a review of bugs, security gaps, speed issues and style.
• Bug cause investigator team: Give it a crash or odd behaviour. Get several parallel read-only investigations and one ranked list of likely causes.
• Bug diagnoser: Describe what is broken or slow. Get the cause found step by step and the fix checked.
• Bug fix evidence recorder: Share a fix and its checks. Get a dated record of what is verified, partial or blocked.
• Bug investigation planner: Paste the bug, the error and the code near it. Get ranked suspects and a check for each before you change anything.
• Bug reproduction evidence writer: Share an unclear bug report. Get repeatable steps, a minimal case and the evidence still needed.
• Bug root cause finder: Describe the bug. Get the cause proven with evidence before any code is changed, so fixes stop being guesses.
• C# API compatibility checker: Share your C# API and web client. Get mismatched fields and breaking changes traced to their consumers.
• Change impact checker: Describe a code change. Get what it could break outside its own diff, and one proof it is safe.
• Change test prover: Give it a Python change. Get tests that fail whenever that change is broken, proven by code mutations.
• CI failure fixer: Point it at a failing pull request. Get the cause of the failed GitHub check and a fix plan.
• Code and documentation consistency checker: Share a feature or code change. Get outdated docs and missing instructions traced to the changed behavior.
• Code change reviewer: Point it at a set of changes. Get a review against your coding standards and your spec, side by side.
• Code change test coverage review: Give it a code change and your tests. Get each new behaviour marked covered or not, plus test cases for the gaps.
• Code cleanup planner: Point it at messy code. Get the problem spots named and cleaned up with proven patterns.
• Code correctness planner: Give it a tricky function. It writes the rules and edge cases first, then code that keeps them.
• Code habits reviewer: Give it code. Get duplication, tight coupling and weak contracts flagged, with fixes. Uses Pragmatic Programmer principles.
• Code refactor planner: Give it messy code. Get named refactoring moves to apply in small safe steps, without changing what it does.
• Code simplifier: Point it at code that works but is hard to read. Get a simpler version that behaves the same.
• Cypress flaky test fixer: Share your Cypress suite. Get fragile waits and old APIs found, then fixes for the affected tests.
• Database migration conflict fixer: Describe the Drizzle migration clash. Get it diagnosed and repaired, with steps to stop it coming back.
• Desktop app tester: Point it at your Electron app. Get it launched and driven by clicks and typing, with screenshots and logs.
• Diagnose slow Qdrant search: Share query traces and workload sizes. Get speed fixes, capacity choices and health measures to compare.
• Django speed reviewer: Point it at a Django project. Get slow queries and N+1 problems found, with the fix for each.
• Documentation checker: Give it a README or doc. Get false claims, drift from the code and weak spots listed before you publish.
• Engineering change verification agent: Share a change and prior review state. Get fresh test and docs gaps, resolved findings and the next verification step.
• Enzyme test migration helper: Share your Enzyme tests. Get React Testing Library replacements suited to a React 18 upgrade.
• Error and log debugger: Give it an error, log or stack trace. Get the cause found and a fix that works.
• Failed test repair agent: Point it at a red test suite. Get capped rounds of fixes, re-checked each time, until green or stopped.
• Failed workflow diagnoser: Paste a failed GitHub Actions log. Get the root cause and a fix for your workflow file.
• Failing test fixer: Point it at a broken test suite. Get failures grouped by cause and fixed one group at a time.
• Failure drill planner: Name a system. Get safe experiments that break parts on purpose, with limits, checks and lessons.
• Fake data finder: Give it a live page. Get a report of which numbers are real, mocked, hardcoded or broken.
• Famous engineer code reviewer: Give it code. Get blunt reviews in the voices of well-known engineers, each tied to their written views.
• Feature fixer: Name the broken feature. Get it traced end to end, fixed and tested without touching the rest.
• Find what is slow, fix it and show the numbers: Give it what feels slow. Get the cause ranked by cost, a fix, and before and after timings.
• Fix a bug with proof and a clear pull request: Give it a bug. Get the cause proven, a small fix, missing tests found and a pull request.
• Fix loop breaker: Give it a bug your AI keeps failing to fix. Get a three-try repair plan, proof it works and a way back.
• Fixed-point code reviewer: Give it MATLAB fixed-point code. Get slow or risky patterns found, with cleaner replacements.
• Flaky test fixer: Give it a failing or flaky Playwright test. Get the cause found and the fix made.
• Fragile code spotter: Paste code an AI wrote fast. Get the few real risks that will hurt later and the smallest fix for each.
• Generated code reviewer: Give it code your AI just wrote. Get a pass for clean-code slips, repeated logic and common AI mistakes.
• GPU code speed tuner: Give it MATLAB code bound for the GPU. Get faster generated CUDA code, measured after each change.
• Hidden bug hunter: Point it at your code. Get a hard line-by-line hunt for logic errors and gaps, sorted by how bad they are.
• Hostile code reviewer: Point it at recent changes. Get a tough review from three hostile reviewers, with each finding ranked.
• Human review gate: Finish AI work. Get a one-page review you sign off before anything counts as done, with proof it happened.
• Library change finder: Point it at a Python project. Get which library changes came after the AI's training and where your code uses them.
• Lint and type check runner: Run it after a change. It runs your project's checks, fixes real failures and reports what did not run.
• Live code debugger: Point it at a bug. Get the program run with breakpoints, so you see exact values and where it goes wrong.
• Live screen debugger: Describe the screen bug. Get your web or Electron app opened, tried out and fixed step by step.
• MATLAB bug tracer: Give it code that fails. Get the cause found, with breakpoints and variable checks to show it.
• MATLAB code reviewer: Give it MATLAB code. Get a review of style, speed and upkeep, checked against MathWorks standards.
• MATLAB memory saver: Give it code that runs out of memory. Get the cause found, fixed and measured, with a short report.
• MATLAB speed test writer: Give it MATLAB code to time. Get repeatable performance tests that measure run time with sound statistics.
• MATLAB speed tuner: Give it slow code. Get the real slowdown found, fixed and measured, with no made-up speed claims.
• MATLAB test runner: Point it at your tests. Get them run and filtered, with failures, coverage gaps and CI set up.
• MATLAB unit test writer: Give it MATLAB code. Get class-based unit tests with fixtures, mocks and app tests that you can run.
• MCP server test runner: Point it at your MCP server. Get realistic task tests and a report on whether an AI picks the right tool.
• MEX file upgrader: Give it old C, C++ or Fortran MEX files. Get them moved to the newer complex number format.
• Mistake-proofing designer: Point it at code, a form or a process. Get changes that make the wrong move impossible or loud.
• Model math debugger: Give it the PyTensor code that errors or runs slow. Get it fixed, with checks that the numbers still match.
• Old code paydown guide: Give it a large old codebase. Get a safe plan to make it legible, bounded and resilient, step by step.
• Old MATLAB code updater: Give it code that uses removed or discouraged functions. Get current replacements in place.
• Open-source contribution preparer: Point it at an open-source issue. Get a checked fix, rerun tests and a drafted post held for your approval.
• Parallel code reviewer: Point it at a code change. Get four reviewers working at once and one short list of issues that matter.
• Parfor error fixer: Paste a failing parfor loop. Get the cause of the error and a fixed loop that runs in parallel.
• Pester test upgrade helper: Share your PowerShell tests. Get changes for the next Pester version and checks for lost test behavior.
• PowerShell pitfall checker: Give it a Windows script. Get operator, quoting and error handling mistakes caught and fixed.
• Pre-commit hook setter: Point it at your repo. Get checks for formatting, types and tests that run before every commit.
• Production error summarizer: Ask about recent errors. Get the Sentry issues and events summed up with the key details.
• Prompt change tester: Give it a prompt change and a fixed set of cases. Get a clear report of what got better or worse.
• Prototype to production guide: Give it a working prototype. Get tests, cleaner structure and a plan to make it ready for real users.
• Pull request reviewer: Point it at each new pull request. Get a risk rating, ranked findings and a review to post once you say yes.
• Pull request risk reviewer: Give it a pull request. Get issues ranked by severity, security notes and clear review comments.
• Pull request risk scorer: Point it at a pull request. Get complexity and risk scored, code smells flagged and review comments written.
• QA test scenario writer: Paste your user stories. Get test scenarios with starting state, steps and the result to expect.
• Quality bar setter: Answer a few questions. Get a written quality bar for your project and a watch for anyone lowering it.
• Quarkus integration test builder: Share your Quarkus app. Get tests with local service containers and checks for real API behavior.
• React 19 source migration helper: Share your React code. Get source changes for React 19 APIs, refs and context.
• React 19 test migration helper: Share your React tests. Get updates for React 19 test APIs and render behavior.
• React batching regression fixer: Share a React class component. Get fixes for state updates that break under automatic batching.
• React legacy context migrator: Share old React context code. Get a migration to modern context with the affected consumers checked.
• React string ref migrator: Share components that use string refs. Get modern refs with their call sites updated.
• React test writer: Share your React app. Get tests for components, linked features and full user tasks that lack coverage.
• React unsafe lifecycle migrator: Share your React class components. Get replacements for unsafe lifecycle methods and notes on changed behavior.
• Readable code reviewer: Give it code. Get clearer names, smaller functions and tidier error handling, each with a reason. Uses Clean Code.
• README first-run tester: Point it at a project. Get each README step marked working or broken, with proof, and the edits that fix it.
• Review a pull request, fix the comments and sign off: Give it an open pull request. Get risks ranked, comments fixed in code and a signed review.
• Review comment fixer: Point it at your open pull request. Get each reviewer comment handled and the code updated.
• Sample data file maker: Say which columns and rules you need. Get a realistic fake dataset as CSV, JSON, SQL or a script.
• Schedule timing checker: Give it a cron schedule. Get a plain-English meaning, the next run times, and any silent mistakes that stop or repeat it.
• Shared field change checker: Name a changed field or flag. Get the places it is missing or read differently, across database, API and apps.
• Single-cell RNA quality checker: Give it an h5ad or h5 file. Get low-quality cells filtered and plots showing data quality.
• Slow code finder: Point it at slow code. Get CPU, memory and I/O hot spots found, with flame graphs and fixes.
• Slow code fixer: Describe what feels slow. Get the bottleneck found, fixed, and proven faster with before and after numbers.
• Slow code speeder: Give it slow code. Get a faster version, one change at a time, with tests run before and after.
• Slow query fixer: Give it a slow SQL query. Get the cause found and a faster version with indexes where needed.
• Slow system investigator: Describe a slow or failing system. Get a checked list of causes and a clear incident report.
• Spark job tuner: Give it a slow Spark job. Get changes to partitions, caching and memory that speed it up.
• Speed tuner: Name what feels slow. Get the real bottleneck measured, fixed and measured again.
• SQL Server migration data checker: Share two SQL Server tables. Get schema and row differences to check a data migration.
• Test coverage gap investigator: Share your repo or change. Get missing, weak and stale tests ranked by the behavior at risk.
• Test data seeder: Describe your tables. Get realistic, repeatable test data loaded for development and staging.
• Test framework migrator: Give it Selenium, Playwright, Puppeteer or Cypress tests. Get them converted to the framework you choose.
• Test plan writer: Describe what you are building. Get a test plan with what to test, at what level and where the gaps are.
• Test suite auditor: Point it at a Python project. Get a list of weak, flaky, slow and useless tests, checked by mutation.
• Test suite writer: Share your code. Get tests for units, linked parts and full user tasks, plus a coverage report.
• Test type fixer: Point it at your tests. Get unsafe type casts swapped for shoehorn so test data stays typed.
• Test-first coach: Say what to build or fix. Get it done test first, in small steps from failing test to clean code.
• TestRail test syncer: Give it your TestRail project. Get its test cases turned into Playwright tests, and results sent back.
• Tuning experiment runner: Name the setting to tune and the score to improve. Get a small tested search with a trial log and stop rule.
• Turn a bug report into proof: Share a bug report and repo. Get a repeatable case, missing tests and a record of what is verified.
• Two-model pull request reviewer: Give it a pull request. Get a review where two AI models argue over the code and agree on what to fix.
• Untested code change guide: Give it code with no tests. Get safe steps to add tests and change it without breaking anything.
• Vibe-coded app cleaner: Point it at a fast-built app. Get dead code and broken imports removed without breaking routes or APIs.
• Vue and Pinia test writer: Share your Vue components and stores. Get focused tests for user behavior, state and failure cases.
• Web app tester: Point it at your local app. Get its screens tried in a real browser, with screenshots and error logs.
• Web page control tester: Point it at a page you just built. Get a report of buttons and controls that do not work.
• Website test runner: Share a website and test depth. Get browser checks and a report of problems found.
• WooCommerce code checker: Give it new WooCommerce code. Get common AI mistakes caught, like broken order storage and missed checkout checks.
• WordPress code checker: Give it a new WordPress plugin or theme. Get security, translation and speed problems found before you ship.