Coding standards, frameworks, refactoring, and code review (16.7K skills)
Review and improve code changes on the current branch using Pipecat conventions. Refine names and structure, add type hints and documentation, check example patterns, and make conservative performance improvements while preserving compatibility.
Combine all service research reports for a selected date into a local Markdown digest. Add up to five highlights to help maintainers prioritize updates, review proposed changes, and spot research errors.
Add or improve Google-style docstrings, the explanations embedded in Python code, for modules and classes. Document constructor arguments, public methods, configuration fields, and enum values while preserving existing documentation that already meets the project’s conventions.
Research new models and API capabilities offered by providers behind Pipecat services. Produce dated reports for each service and local committed branches for clear updates, with review commands and separate publishing steps.
Create changelog files for significant changes in a pull request, a proposed code change. Describe effects users will notice and organize entries by PR number and change type, such as features or fixes.
Draft or update GitHub pull request descriptions using the actual changes and related context. Explain the problem, changed behavior, and review concerns in plain language, including breaking changes and explicit issue-closing references when applicable.
Review GitHub pull requests with specialized agents that check for clear bugs and violations of project instructions in CLAUDE.md. Validate bug findings before posting focused comments on the affected lines, with fix suggestions where appropriate.
Create or update Codex skills that package specialized knowledge, repeatable workflows, and tool instructions. The guide covers concrete use cases, folder structure, reusable resources, starter generation, and basic validation of names and required metadata.
Build MCP servers, which let AI agents use external APIs and services through tools. The skill guides tool design, authentication, input validation, implementation, testing, and evaluation using TypeScript or Python.
Review a specified code change without editing files. Checks the full diff, surrounding code, tests, and callers to identify actionable defects introduced by the change, then reports findings by severity with file and line references.
Build micro-frontends with qiankun 3.x, combining separate web apps inside a host app. The skill covers creating host and child apps, converting existing Vite apps, and connecting loading behavior in React or Vue.
Find the right workflow for MNN CPU work, from investigating slow execution to implementing calculation routines and planning tests. Compare ARM, x86_64, and RISC-V backend structures before choosing the relevant guidance.
Improve MNN inference on OpenCL by measuring bottlenecks and refining GPU kernels, the code that performs calculations. The skill guides baseline measurements, correctness checks, Android device benchmarks, and integration of new OpenCL features with compatibility fallbacks.
Triage incorrect MNN outputs and regressions by matching symptoms to focused debugging guides. Use backend comparisons, minimal reproductions, and controlled experiments to investigate memory issues, quantization, concurrency, numeric limits, caches, and inconsistent results across runs.
Add new operators to MNN through a staged workflow covering data definitions, output shapes, computation, and tests. It prioritizes combining existing operators, with CPU implementations and additional hardware backends used when needed.
Improve MNN's Vulkan backend by separating CPU scheduling bottlenecks from GPU computation costs. It guides operator and shader changes, new feature integration with compatibility fallbacks, and device testing to verify both correctness and end-to-end performance.
Optimize and troubleshoot MNN's Hexagon DSP backend, which runs computation on a dedicated processor. The workflow uses device measurements to identify bottlenecks, compare v79 and v81 architectures, verify correctness, and investigate crashes or memory usage.
Diagnose errors and incorrect inference results in MNN's QNN backend for Qualcomm HTP processors. Compare intermediate calculation data against CPU results to locate the first faulty operator, distinguish bugs from precision limits, and guide operator support and performance fixes.
Guide MNN Metal backend development and optimization, from finding performance bottlenecks to changing GPU computation, combining operators, and adjusting execution scheduling. A task-based document index directs work toward relevant implementation rules, benchmarks, and correctness checks.
Develop or port CPU kernels, the low-level calculation routines used by MNN. Follow staged implementation guidance, check data layout agreements, register functions, and verify correctness across processor instruction sets and numerical precision settings.
Find why MNN CPU inference is slow by examining threads, scheduling, memory layout, and the processor instruction path actually used. Operator-level measurements and comparisons with other inference frameworks help identify which layer needs improvement across ARM, x86_64, and RISC-V.
Walks you through an Open SaaS project, explaining its structure, features, and customization checklist. It connects the documentation to relevant project files and code, pausing after major sections so you can ask questions.
Guides you through installing Wasp agent skills, which add Wasp knowledge and recommended practices to AI coding tools. It retrieves installation instructions and explains the steps for working with Wasp and Open SaaS.
Helps you start an Open SaaS project by checking Wasp installation and guiding setup for your operating system. Using the Getting Started guide, it assists with database startup, database migrations, and running the development server.