Claude Skills by TerminalSkills

Deploy and configure Onyx, a self-hosted AI chat platform with RAG and 25+ data connectors. Use when: setting up private ChatGPT alternative, connecting AI to company documents, building enterprise AI assistant with Confluence/Slack/Drive access.

Expert guidance for OPA (Open Policy Agent), the CNCF policy engine for unified authorization across the stack. Helps developers write Rego policies for Kubernetes admission control, API authorization, infrastructure-as-code validation, and data filtering — enforcing security policies as code.

Open SWE is LangChain's open-source asynchronous coding agent: a self-hosted service that takes tasks from a dashboard, GitHub, Slack or Linear, works in an isolated sandbox and opens pull requests. Use when a user asks to deploy or run Open SWE, set up its GitHub App and Slack bot, trigger it with @openswe, have it review pull requests, call it from CI with the oswe CLI, or customize its model, sandbox and prompts.

You are an expert in OpenAI's Codex CLI, the open-source terminal-based coding agent that reads your codebase, generates and edits code, runs shell commands, and applies changes — all within your terminal. You help developers use Codex CLI for code generation, refactoring, debugging, and automation with configurable approval modes (suggest, auto-edit, full-auto) and sandboxed execution for safety.

Build voice-enabled AI applications with the OpenAI Realtime API. Use when a user asks to implement real-time voice conversations, stream audio with WebSockets, build voice assistants, or integrate OpenAI audio capabilities.

Design and maintain OpenAPI specifications. Use when a user asks to create an API spec, document REST APIs, generate client SDKs from a spec, validate API contracts, or implement design-first API development.

OpenBB Platform is an open-source Python library, REST API and MCP server that gives one interface to financial data: stocks, ETFs, crypto, forex, macro economics and SEC filings. Use when building financial analysis tools, feeding market data to AI agents, creating quantitative research pipelines, or accessing free financial data APIs from Python.

Deploy and manage OpenClaw, a self-hosted gateway bridging messaging platforms to AI coding agents. Use when a user asks to set up OpenClaw, connect WhatsApp or Telegram or Discord to an AI agent, configure multi-agent routing, schedule cron jobs in OpenClaw, set up webhooks, manage OpenClaw channels, pair a messaging account, configure heartbeats, spawn sub-agents, or troubleshoot OpenClaw gateway issues. Covers installation, channel setup, agent configuration, cron scheduling, webhooks, and...

You are an expert in OpenCV (Open Source Computer Vision Library), the most popular library for real-time computer vision. You help developers build image processing pipelines, object detection systems, video analysis tools, augmented reality, and document processing using OpenCV's 2,500+ algorithms for image manipulation, feature detection, camera calibration, 3D reconstruction, and DNN inference — in Python, C++, or JavaScript.

OpenDataLoader PDF converts PDF files into Markdown, JSON with bounding boxes, HTML or plain text, and can add accessibility tags to untagged PDFs. It runs locally on Java with a CLI and Python, Node.js and Java APIs. Use when a user asks to parse PDFs for RAG, convert a PDF to Markdown, extract tables or headings with page coordinates, OCR scanned PDFs, or produce Tagged PDFs.

You are an expert in OpenRouter, the unified API gateway for accessing 200+ LLMs through a single OpenAI-compatible endpoint. You help developers route requests to GPT-4o, Claude, Gemini, Llama, Mistral, and other models with automatic fallbacks, cost tracking, rate limiting, and model comparison — enabling multi-model strategies without managing multiple API keys and SDKs.

Create stunning screen recordings and product demos with OpenScreen — open-source, no watermarks, free for commercial use. Use when: recording product demos, creating tutorial videos, building marketing content, screen recording with post-processing effects.

You are an expert in OpenTofu, the open-source fork of Terraform maintained by the Linux Foundation. You help developers and platform teams provision cloud infrastructure using HCL (HashiCorp Configuration Language), with full compatibility with existing Terraform modules, state files, and providers — plus new features like client-side state encryption, OCI registry support, and removed BSL license restrictions.

Parse and manipulate OpenType/TrueType fonts with opentype.js — read font metadata, access glyph outlines, measure text, generate font subsets, and render text to SVG paths. Use when tasks involve custom text rendering, font analysis, glyph extraction, or building font tools.

Manage AI agent context (memory, resources, skills) using OpenViking's file system paradigm. Use when: building agents with persistent context, managing agent memories across sessions, implementing hierarchical context delivery for complex agent systems.

OpenVPN is an open-source VPN daemon that builds encrypted TLS tunnels for remote access and site-to-site links. Use when a user asks to set up a VPN server, create client certificates, configure site-to-site tunnels, set up split tunneling, manage PKI with EasyRSA, harden OpenVPN security, automate client provisioning, configure routing and NAT, set up MFA for VPN, monitor connected clients, revoke a client, or troubleshoot VPN connectivity.

Configure Opsgenie for alert management, on-call scheduling, routing rules, and incident response. Use when a user needs to set up alert routing, create escalation policies, manage on-call rotations, integrate with monitoring tools, or automate incident workflows with Opsgenie.

Add terminal spinners with Ora. Use when showing loading indicators in CLI tools, displaying progress during async operations, or building polished command-line interfaces with status feedback.

Orama is an in-process search engine for JavaScript and TypeScript: full-text, vector and hybrid search that runs in the browser, on a server or at the edge with no external service. Use when someone asks to "add search to my docs site", "client-side search", "typo-tolerant search", "faceted search in JavaScript", "vector search without a database", or mentions Orama or @orama/orama. Covers schemas, filters, facets, sorting, embeddings, React, and saving an index to a file.

Runs a large refactor or migration across many files by splitting it into units that cannot collide, mapping the code with read-only subagents, dispatching one worker per unit in dependency order, and verifying everything in the main session before reporting. Includes the decision of when not to parallelize, a plan format with file ownership, a worker brief, and how subagents and worktrees behave in Claude Code, Codex, Gemini CLI and Cursor. Use when the user says "refactor this across the wh...

Calls the Originality.ai API to score text for AI authorship and to check it for plagiarism. Use when auditing SEO content, verifying freelancer submissions, building an editorial review queue, or when a user asks to "check if this article is AI-written" or "run an originality check".

Instrument Node.js apps with OpenTelemetry. Use when a user asks to add distributed tracing, collect metrics, instrument HTTP requests, trace database queries, or set up observability for microservices.

Generates guaranteed-valid structured text from LLMs with Outlines, the Python library that constrains token sampling to a JSON schema, regex, choice list or grammar. Use when a user asks for reliable JSON from a local or hosted model, classification into fixed labels, regex-shaped output, grammar-constrained generation, or structured output with vLLM, Transformers, Ollama or OpenAI-compatible servers.

Send, read, and manage emails via Microsoft Outlook using the Graph API. Use when someone asks to "send email via Outlook", "read Outlook inbox", "automate Outlook emails", "manage Outlook folders", "Outlook API integration", "email automation with Graph API", "search Outlook emails", or "process incoming emails". Covers sending (plain, HTML, attachments), reading, searching, folders, rules, and focused inbox via Microsoft Graph API.

OWASP ZAP (Zed Attack Proxy) is an open-source web application security scanner that crawls a site, passively inspects traffic and optionally attacks it to find XSS, SQL injection, missing security headers and other OWASP Top 10 issues. Use when configuring automated scans, running zap-baseline, zap-full-scan or zap-api-scan, adding security scanning to CI/CD or GitHub Actions, or triaging scan results. Trigger words: owasp zap, security scan, vulnerability scanner, penetration testing, zap-b...

Lint JavaScript/TypeScript at blazing speed with oxlint — a Rust-based linter 50-100x faster than ESLint. Use when someone asks to "speed up linting", "oxlint", "fast JavaScript linter", "replace ESLint", "Rust linter for JS", or "lint in CI faster". Covers rule configuration, ESLint migration, framework plugins, and CI integration.

HashiCorp Packer builds identical machine images (AWS AMIs, Google Compute images, Docker images and more) from one HCL template, with provisioners that configure the image and plugins for each platform. Use when the user wants to bake an AMI or golden image, build a Docker image with provisioners, write a .pkr.hcl template, or fix packer init, validate and build errors.

When the user wants to implement consumer-driven contract testing between microservices using Pact. Also use when the user mentions "pact," "contract testing," "consumer-driven contracts," "CDC testing," "provider verification," or "Pact Broker." For API mocking, see mockoon or wiremock.

Paddle is a merchant-of-record payments platform that sells software on a company's behalf and handles checkout, subscriptions, sales tax, VAT and invoicing. Use when a user asks to add subscription billing without handling tax compliance, accept international payments, integrate Paddle Checkout or Paddle webhooks, build a SaaS billing system on Paddle Billing, or let customers manage and cancel subscriptions.

Audits a marketing page (homepage, ad landing page, pricing, feature page or article) to find why visitors do not take the next step, and returns ranked fixes with evidence plus a test plan sized for the page's traffic. Use when someone says "this page isn't converting", "audit my landing page", "improve conversions", "CRO", "conversion rate optimization", "why is my bounce rate high", or "what should we A/B test". Measures by device and source first, captures what the visitor actually reads,...

Configure PagerDuty for incident management, on-call scheduling, alert routing, and escalation policies. Use when a user needs to set up PagerDuty services, create escalation policies, configure integrations with monitoring tools, manage on-call rotations, or automate incident workflows.

Plans, structures, launches and diagnoses paid campaigns on Google Ads, Meta (Facebook and Instagram) and LinkedIn against a cost-per-acquisition or return target derived from the business's own numbers. Use when someone asks to "set up Google Ads", "run Meta ads", "write ad copy", "lower our CPA", "improve ROAS", "plan a PPC budget", "set up retargeting", or "why are my ads not converting". Covers the budget check, conversion tracking, campaign structure, bidding, ad text within platform lim...

Assists with loading, cleaning, transforming, and analyzing tabular data using pandas. Use when importing CSV/Excel/SQL data, handling missing values, performing groupby aggregations, merging datasets, working with time series, or building analysis-ready datasets. Trigger words: pandas, dataframe, csv, groupby, merge, time series, data cleaning.

Pandera is a Python library for validating DataFrames (pandas, Polars, PySpark, Dask and others) against declarative schemas with column types, constraints and custom checks. Use this skill when asked to validate a DataFrame, define a data contract, add data quality checks to a pipeline or pytest suite, or report which rows fail validation.

Build multiplayer and collaborative apps with PartyKit. Use when a user asks to build real-time collaboration features, create multiplayer experiences, add live cursors, build collaborative editing, create real-time voting or polls, sync state across clients, or build WebSocket servers on the edge. Covers PartyServer definition, client connections, state management, hibernation, and deployment to Cloudflare.

Build and extend 3D building editor apps with Pascal Editor, an open-source React Three Fiber and WebGPU editor whose scene is a flat Zustand node store (site, building, level, wall, door, window, slab, zone). Use when building a 3D architectural tool or BIM-like editor, generating floor plans from code or an AI agent, extending Pascal with custom node plugins, or running its MCP server.

Comprehensive passive reconnaissance covering WHOIS, DNS history, reverse IP, SSL certificates, Google dorks, and Wayback Machine enumeration. Use when: initial target profiling without active scanning, stealth recon where no direct target contact is desired, or building a complete picture of a target's infrastructure from public data alone.

Passport.js is authentication middleware for Express and other Node.js frameworks, with strategies for local passwords, Google, GitHub and many other providers. Use when implementing login with Google/GitHub/email, OAuth, session auth, or social login.

Payload is an open-source, code-first headless CMS and application framework that runs inside a Next.js app, with collections defined in TypeScript. Use this skill when defining collections in TypeScript, configuring access control, customizing the admin panel, or integrating with Next.js. Trigger words: payload, payload cms, headless cms, collections, admin panel, content management, payload fields.

Designs and audits the moments inside a product where an account is asked to pay or to move to a higher plan: feature gates, usage-limit screens, trial-ending and trial-expired states, seat and add-on upsells. Maps every upgrade moment, decides what blocks and what only warns, specifies each screen with the exact charge, wires checkout and entitlements, and sets the events and guardrails to judge it. Use when someone says "paywall", "upgrade modal", "feature gate", "limit reached screen", "tr...

Achieve PCI DSS compliance for handling payment card data — scoping, controls, and SAQ selection. Use when storing, processing, or transmitting payment card data, building e-commerce security, or achieving payment processing compliance.

Extract text, tables, metadata, and structured data from PDF files. Use when a user asks to read a PDF, parse a PDF, extract data from a PDF, summarize a PDF document, pull tables from a PDF, or convert PDF content to structured formats like JSON or CSV. Handles single and multi-page documents, scanned PDFs, and PDFs with complex table layouts.

Combine multiple PDFs into one or split a PDF into separate files. Use when a user asks to merge PDFs, combine PDF files, join documents together, split a PDF into pages, extract pages from a PDF, or separate a PDF into parts. Supports page range selection and custom ordering.

Extract text from scanned PDFs using optical character recognition. Use when a user asks to OCR a PDF, read a scanned document, extract text from an image PDF, digitize a scanned file, convert a scanned PDF to text, or read text from a photograph of a document. Supports multiple languages and handles low-quality scans.

Generate PDF documents programmatically with PDFKit — add text, images, vector graphics, tables, headers/footers, and custom fonts. Use when tasks involve creating invoices, reports, certificates, or any structured PDF output from application data in Node.js.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when a user asks to fine-tune a language model, train a custom LLM, adapt a model to their data, use LoRA or QLoRA, fine-tune Llama or Mistral, or train a model on consumer GPUs. Covers PEFT methods for 7B-70B parameter models.

PentAGI is a self-hosted, autonomous AI penetration-testing system: LLM agents plan and run security tools such as nmap and sqlmap inside sandboxed Docker containers and write a vulnerability report. Use when a user asks to automate security testing, set up autonomous pentesting, deploy an AI-driven vulnerability scanner, build a self-hosted security testing platform, or drive pentests through the PentAGI API. Only for systems the user is authorized to test.

When the user wants to perform visual testing with automated screenshot comparison using Percy by BrowserStack. Also use when the user mentions "percy," "visual testing," "screenshot comparison," "visual diff," "percy snapshot," or "BrowserStack visual." For Storybook-specific visual testing, see chromatic.

Review code for performance issues and optimization opportunities. Use when someone needs to check for N+1 queries, unnecessary re-renders, memory leaks, inefficient algorithms, missing indexes, or bundle size regressions. Trigger words: performance review, slow query, N+1, memory leak, bundle size, latency, optimization, re-render, Big O.

Perplexity API gives LLM answers grounded in live web search, with cited sources. Use when you need up-to-date information in AI responses, research assistants, fact-checking, current events coverage, or any task requiring knowledge beyond an LLM's training cutoff. Returns cited sources alongside answers.