Claude Skills by agentskillexchange

Use csv-diff when an agent needs to explain what changed between two structured exports, not just that the files differ. The agent lines records up by a stable key, reports added, removed, and changed rows, and can hand the result to humans or downstream automations as readable text or machine-friendly JSON.

Resolve Python dependency inputs into deterministic lock files and sync environments without hand-editing transitive pins.

Tracks cyclomatic and cognitive complexity trends using lizard CLI and radon for Python analysis. Generates weekly complexity reports with git-integrated change attribution per module.

Connect LLMs to MCP servers through composable patterns like router, evaluator-optimizer, and orchestrator flows without hand-managing server lifecycles.

Use Lobster when an OpenClaw operator wants one deterministic typed workflow step, with approval gates and resumable execution, instead of re-planning the same multi-step tool sequence in chat.

Composio provides 1000+ pre-built toolkits with managed authentication, context management, and sandboxed execution to connect AI agents to external apps like Gmail, Slack, GitHub, and Notion. Available as Python and TypeScript SDKs with support for major agent frameworks.

Reduces token burn by compressing command output and deduplicating repeated file reads before they are sent to Claude Code, Codex, Cursor, and related coding-agent tools.

Concurrently runs multiple commands in parallel with color-coded output, prefix labels, and process lifecycle management. With 14M+ weekly npm downloads, it is the standard tool for running dev servers, watchers, and build processes simultaneously.

Attach judges to LaunchDarkly AI Config variations, create custom judges, set sampling rates, and interpret production quality signals from online evaluations.

Set up Craft Agents workspaces with connected sources, imported skills, MCP servers, permission modes, and session workflow states for supervised agent work.

Conftest is a CLI tool that tests structured configuration data using the Open Policy Agent Rego language. It validates Kubernetes manifests, Terraform configs, Dockerfiles, CI pipelines, and any other structured format against custom policy rules.

Basic usage or getting-started notes: - Here's a quick example. Save the following as policy/deployment.rego: - Assuming you have a Kubernetes deployment in deployment.yaml you can run Conftest like so:

Use Azure DevOps MCP when an agent needs governed access to ADO projects, work items, builds, repos, test plans, wikis, and pipelines.

Give Claude Code local command access to Reminders, Calendar, Contacts, Notes, Mail, and tmux for workflow-aware personal ops on macOS.

Use BatchData property and address APIs from MCP clients to enrich real-estate research and operations workflows.

Expose Zendesk tickets, comments, and Help Center context to MCP-capable agents for supervised support workflow preparation.

Contact Form 7 is the long-running WordPress forms plugin from Rock Lobster Inc., with more than 10 million active installs and official documentation at contactform7.com. This skill is for agents that need to understand how a site’s forms, mail templates, validation behavior, and shortcode placements are actually handled in Contact Form 7.

Scans Docker and OCI container images for vulnerabilities using Trivy JSON output and the Docker Hub API v2 for image metadata. Analyzes base image layers via Syft SBOM generation and maps CVEs to fixed versions using the Alpine SecDB and Debian Security Tracker APIs.

Monitors container runtime behavior using Falco rules and the Docker Engine API. Detects anomalous syscalls, privilege escalations, and unexpected network connections in real time.

Creates detailed SEO content briefs using Ahrefs API keyword data, SurferSEO NLP analysis, and Clearscope content grading. Generates outlines with target word counts, headers, and semantic keyword clusters.

Identifies keyword cannibalization across site pages by analyzing Google Search Console Performance API data, extracting query-page pairings, and detecting overlap using TF-IDF cosine similarity scoring. Generates merge recommendations and 301 redirect maps via Screaming Frog exports.

Scores content using Flesch-Kincaid, Gunning Fog, and SMOG readability indices via textstat Python library. Provides sentence-level rewrite suggestions using OpenAI GPT-4o API.

Analyzes and optimizes content readability using textstat Python library and Hemingway API patterns. Computes Flesch-Kincaid, Gunning Fog, and SMOG indices with automated rewriting suggestions.

Use the Contentful Management API and JavaScript SDK to create, update, validate, and publish entries, assets, content models, and environments in a headless CMS workflow. This skill gives an agent a concrete path for operating Contentful programmatically rather than treating it like a generic CMS.

Context7 by Upstash injects up-to-date, version-specific library documentation and code examples directly into AI prompts. Eliminates hallucinated APIs and outdated code generation by pulling documentation straight from the source at query time.

Use Browser MCP to let an MCP-capable agent drive a user's browser session for inspection, navigation, form work, and browser-state-aware automation.

Run an on-device Android control loop that inspects screens and performs app actions locally for mobile automation and testing workflows.

Let an agent drive a real logged-in Chrome profile through MCP for authenticated browsing, parallel tab work, and token-efficient page interaction without re-authenticating each run.

Drive real mobile apps from a session-aware CLI so agents can inspect UI state, act deterministically, and collect logs, network traces, and performance evidence during debugging.

Let MCP-compatible agents inspect, debug, deploy, audit, and manage Kubernetes clusters through a controlled kubectl-backed server.

Use har-to-k6 when an agent has recorded browser traffic and needs to turn it into a repeatable k6 script instead of hand-writing one from scratch. The skill is about transforming captured sessions into a starter load-test artifact with validation and export, not about listing k6 or Grafana as products.

Use olmOCR when an agent needs to turn scanned or layout-heavy documents into clean markdown or text before chunking, search, extraction, or citation workflows.

Use Mammoth when an agent needs to turn a .docx file into simple HTML that preserves semantic structure instead of Word-specific styling. This is for ingestion and publishing workflows, not for full document editing or perfect visual fidelity.

Use html-to-text when an agent receives raw HTML from inboxes, support systems, or scraped pages and needs readable plain text before classification, summarization, or indexing. The skill is deliberately bounded to deterministic HTML-to-text conversion, not crawling or summarization.

Use Mailrise to accept ordinary email alerts and fan them out through Apprise-backed notification channels when legacy systems can only speak SMTP.

Convex is an open-source reactive database for app developers. Write pure TypeScript server functions with strong consistency, real-time subscriptions, and automatic caching. Self-hostable or available as a managed cloud service.

Bootstraps new projects from Cookiecutter templates with variable substitution and post-generation hooks. Supports remote template repositories via the cookiecutter CLI and Jinja2 rendering engine.

Bootstraps project repositories from Cookiecutter templates with Jinja2 variable substitution and post-generation hooks. Supports remote template registries and custom extension plugins.

Creates project scaffolds from Cookiecutter templates with interactive variable resolution. Integrates with cruft for template drift detection and copier for advanced multi-template composition.

Generates project scaffolds from Cookiecutter templates with Jinja2 variable interpolation. Supports post-generation hooks, conditional file inclusion, and template composition from multiple sources.

Generates project boilerplate from Cookiecutter templates with Jinja2 variable injection. Supports custom hooks for post-generation linting via Ruff and pre-commit framework setup.

Scaffolds project templates using the Cookiecutter engine with Jinja2 templating and pre/post generation hooks. Supports conditional file inclusion, directory renaming, and cookiecutter.json variable validation with JSON Schema.

Generates project boilerplate from Cookiecutter templates with interactive variable prompts. Supports custom Jinja2 hooks, post-generation scripts, and template composition from multiple cookiecutter.json sources for complex monorepo scaffolding.

Manages and instantiates Cookiecutter project templates with dynamic Jinja2 variable substitution and post-generation hooks. Supports template inheritance chains and integrates with cruft for template update tracking.

Generates project scaffolds using Cookiecutter templates from GitHub repositories or local directories. Automates cookiecutter.json variable prompting and post-generation hook execution.

Generates project scaffolding using the Cookiecutter CLI with Jinja2 templating. Supports custom template registries, post-generation hooks via Python scripts, and integration with the GitHub template repository API.

Coolify is an open source self-hosted deployment platform for applications, databases, and services. This skill covers installing Coolify, connecting infrastructure over SSH, and using it as an operational control plane for repeatable app delivery workflows.

Run manager-worker agent collaboration in Matrix rooms where humans can watch, intervene, and keep credentials out of worker hands.

Use CCPM when an agent team needs one issue-backed workflow that turns plans into GitHub issues, isolates execution in worktrees, and keeps parallel coding runs reviewable instead of relying on ad hoc chat memory.

Use Roo Code modes to plan, implement, debug, and review coding-agent work from inside VS Code while keeping edits, commands, model choices, and MCP tool use visible to the operator.