Claude Skills by agentskillexchange

Run coding agents inside a repo with explicit policy files, task state, approvals, and closure records when teams need auditable local workflows instead of opaque agent runs.

pyannote.audio is an open-source Python toolkit for speaker diarization built on PyTorch. It provides state-of-the-art pretrained models and pipelines for speech activity detection, speaker segmentation, overlapped speech detection, and speaker embedding.

PydanticAI is an AI agent framework built by the Pydantic team that brings FastAPI-style ergonomics to GenAI development. It provides type-safe, model-agnostic agent construction with structured outputs, dependency injection, and seamless integration with Pydantic Logfire for observability.

pydub is a Python library that provides a simple, high-level interface for manipulating audio files. It supports slicing, concatenation, volume adjustment, crossfading, format conversion, and effects processing across all formats supported by FFmpeg.

Manages Pylint baseline files for gradual code quality improvement using pylint --output-format=json and the pylint.reporters API. Tracks new violations per commit while suppressing pre-existing issues in legacy code.

Perform automated Python code reviews using Pylint programmatic API and pylint-json2html for report generation. Supports custom checker plugins and per-project rcfile configurations.

Queries the PyPI JSON API and the libraries.io API to analyze Python package metadata, dependency trees, and version histories. Uses pip-audit for vulnerability scanning against the OSV database.

Pyroscope is an open-source continuous profiling platform by Grafana Labs that helps identify CPU, memory, and I/O bottlenecks at the line-of-code level. It integrates natively with Grafana for flame graph visualization and supports profiling for Go, Java, Python, Ruby, Node.js, Rust, and .NET applications.

Requirements and caveats from upstream: - .. code-block:: python - <https://docs.pytest.org/en/stable/explanation/goodpractices.html#python-test-discovery>_ - Python 3.10+ or PyPy3

Resolves and visualizes Python package dependency graphs using pip, pipdeptree, and the PyPI JSON API. Detects version conflicts, circular dependencies, and vulnerable transitive dependencies.

Extracts and indexes Python module documentation using ast.parse and docstring_parser library. Supports Google, NumPy, and Sphinx docstring formats with type hint cross-referencing via typing_inspect.

Validates and auto-generates Python docstrings using the ast module and pydocstyle conventions. Supports NumPy, Google, and Sphinx docstring formats with type stub integration.

Generates searchable Python library references using ast module for source parsing and Sphinx autodoc integration. Extracts docstrings, type hints, and usage examples from installed packages via importlib.

Maps the public API surface of Python packages using ast module parsing and importlib introspection. Generates comprehensive reference docs with type annotations from mypy stubs.

Builds dependency graphs for Python packages using the PyPI JSON API and pipdeptree library. Visualizes transitive dependency chains and identifies version conflict risks.

Retrieves and indexes Python package documentation from PyPI metadata API and Read the Docs API. Uses ast module parsing and pydoc introspection to extract function signatures, docstrings, and type hints for offline reference.

Resolves Python package dependencies using the PyPI JSON API and pip resolver algorithm. Generates locked requirements files and checks compatibility across Python version markers via packaging library.

Indexes Python package documentation using the PyPI JSON API and Read the Docs API. Builds searchable reference catalogs with function signatures, type hints, and usage examples.

Measures and improves Python type annotation coverage using mypy --html-report and pyright type completeness scoring. Identifies untyped function signatures, missing return types, and Any-typed parameters across codebases.

An official Qdrant MCP server implementation that provides semantic memory capabilities for AI agents. Enables storing and retrieving information using vector search, acting as a persistent knowledge layer on top of the Qdrant vector database.

qsv is a high-performance command-line toolkit for querying, transforming, validating, and analyzing CSV and tabular data. Written in Rust, it provides over 50 commands for data wrangling tasks and supports Excel, Parquet, JSON, and other formats.

Connect an MCP client to PostgreSQL databases for schema discovery, safe querying, and database inspection through a documented pgEdge MCP server and agent client workflow.

Use mdq when an agent needs to target headings, lists, links, or other Markdown structure without falling back to brittle regex edits.

Use Dinobase to give agents a SQL layer over SaaS APIs, databases, files, and MCP servers so they can answer cross-system business questions with previewed writes.

Use Fabric RTI MCP when an agent needs tool-callable access to Microsoft Fabric Real-Time Intelligence services such as Eventhouse, Eventstreams, Activator, and Map instead of sending the user back to the Fabric UI or wiring custom SDK glue.

Query Monitor is the developer tools panel for WordPress and WooCommerce. It enables debugging of database queries, PHP errors, hooks and actions, block editor blocks, enqueued scripts and stylesheets, HTTP API calls, and more with filtering by plugin or theme.

Connect MCP-compatible agents to Neo4j so they can inspect graph schemas, run Cypher queries, manage graph memory, and operate Aura instances from chat.

Connect an MCP-compatible assistant to an existing PostgreSQL database for safe natural-language querying, schema-aware SQL, and streamed results.

Use the upstream install or setup path that matches your environment: - npm install kafkajs

Connects to the QuickBooks Online Accounting API using OAuth 2.0 via the intuit-oauth Node.js SDK to fetch unpaid invoices and match them against bank transaction records. Discrepancies are flagged and a reconciliation report is generated as a PDF using PDFKit, then emailed via SendGrid.

Generate strongly-typed models and serializers from JSON, JSON Schema, TypeScript, and GraphQL queries in 20+ programming languages. quicktype turns sample data into production-ready type definitions for Swift, C#, Go, Python, Rust, Java, Kotlin, and more.

Quickwit is a cloud-native search engine built in Rust for log management and distributed tracing. It offers sub-second search on cloud storage (S3, Azure Blob, GCS), an Elasticsearch-compatible API, native OpenTelemetry and Jaeger integration, and decoupled compute and storage architecture.

Radicale is a lightweight, self-hosted CalDAV and CardDAV server written in Python. It shares calendars, to-do lists, journal entries, and contacts over standard protocols with zero-config setup, file-based storage, optional authentication, TLS support, and a plugin architecture.

Guide an agent through property search, buyer/renter preference capture, and evidence-backed shortlist notes from structured listing data.

Basic usage or getting-started notes: - bash - bash <(curl -fsSL railway.com/install.sh) --agents -y - This installs the CLI to ~/.railway/bin and runs [railway setup agent](https://docs.railway.com/cli/setup) to configure detected agent tools.

Rallly is an open-source scheduling and collaboration tool that makes organizing group meetings effortless. Create meeting polls to find the best date and time based on participant availability, replacing back-and-forth email chains with a simple voting interface.

Rank Math SEO is the WordPress SEO plugin from Rank Math SEO, with more than 3 million active installs and current releases published through WordPress.org. This skill helps agents work with Rank Math’s metadata, schema, redirects, content analysis, and search-optimization settings using the plugin’s real configuration model.

Rclone is an open-source command-line program to sync files and directories to and from over 70 cloud storage providers including Google Drive, S3, Dropbox, Azure, OneDrive, and Backblaze B2. Often called the Swiss Army knife of cloud storage, it handles copy, sync, mount, and serve operations with encryption support.

React Email is an open source component library for building responsive, cross-client email templates using React and TypeScript. It handles Gmail, Outlook, and Apple Mail inconsistencies, supports dark mode, and renders to standard HTML for use with any email provider.

Search channels, inspect threads, move files, and send or edit Slack messages through an agent-oriented CLI with structured output.

Connect an MCP-capable assistant to Google Drive search, file reads, and Google Sheets cell updates with mcp-gdrive.

Uses Subaligner to retime an existing subtitle file against the final audio track, then outputs a corrected subtitle asset. This is for subtitle drift, forced alignment, or batch retiming, not for full video editing or general media management.

RealtimeSTT is a Python library for real-time speech-to-text with advanced voice activity detection, wake word activation, and instant transcription. It combines WebRTC VAD, Silero VAD, and Faster Whisper for production-grade voice input in agent applications.

Recon-ng is a full-featured modular reconnaissance framework written in Python, designed to conduct web-based open source intelligence (OSINT) gathering quickly and thoroughly. Its Metasploit-like interface and extensible module system make it the standard tool for structured OSINT workflows.

Use VCR.py when an agent needs to turn flaky, slow, or rate-limited Python tests into stable runs by recording real HTTP interactions once and replaying them from cassette files. The agent decides which requests belong in fixtures, refreshes stale cassettes when upstream APIs change, and keeps external traffic out of the repeat test loop.

Capture a terminal session as a replayable cast when an agent needs a text-native CLI repro or walkthrough instead of a generic screen recording.

Run local adversarial attack passes against agents, RAG pipelines, and chatbots to surface concrete failure classes before production rollout.

Use scrubadub when an agent needs to strip emails, phone numbers, names, and similar sensitive text before sending content to external systems or search indexes.

Uses the Reddit OAuth2 API via PRAW to collect top posts and comments from subreddits on a rolling 24-hour window, then runs batch sentiment scoring via HuggingFace Inference API using twitter-roberta-base-sentiment. Aggregates scores into a daily time-series written to Google Sheets for Looker Studio visualization. Fires Slack alerts when sentiment drops more than 20 points week-over-week.

Use the upstream install or setup path that matches your environment: - docker run -d -p 6379:6379 redis:latest - cmake --version - make -j "$(nproc)" all - brew update