llms.txt
Compare with other sites' llms.txt files →# AgentsKit
> AgentsKit is a paid toolkit for [Claude Code](https://claude.com/claude-code) that drops a full AI engineering and marketing team into any project with one command: 89 specialized agents, 122 pre-built skills, and 181 slash commands. It works with any tech stack, adapts to your conventions via a CLAUDE.md file, and ships as two kits (Engineer + Marketing) you can install separately or together.
AgentsKit is an independent, unofficial product. It is not affiliated with, endorsed by, or sponsored by Anthropic. "Claude" and "Claude Code" are trademarks of Anthropic. It is delivered as a private GitHub repository and installed with a one-line `npx github:getagentskit/… init` command (no npm publish, no global install; Node 18+).
## What you get
- 89 specialized agents (58 engineering + 31 marketing)
- 122 pre-built skills (61 engineering + 61 marketing)
- 181 slash commands (159 engineering + 32 marketing; 10 shared docs commands counted once)
- Two kits: Engineer (the software team) and Marketing (the growth team)
- CLAUDE.md templates so agents learn your stack, conventions, and brand voice
- The `agentskit` installer CLI; cherry-pick any single agent, skill, or command
- Lifetime updates: buy once, run `agentskit update` (or `git pull`) forever
## Kits
- [Engineer kit](https://agentskit.co/#kits): 58 agents, 61 skills, 159 commands. Includes tech-lead, backend-architect, react-specialist, typescript-pro, python-pro, golang-pro, rust-pro, postgres-pro, kubernetes-specialist, terraform-specialist, sre-engineer, code-reviewer, test-automator, debugger, security-auditor, and commands like /api-scaffold, /test-coverage, /deploy-checklist.
- [Marketing kit](https://agentskit.co/#kits): 31 agents, 61 skills, 32 commands. Includes growth-strategist, brand-voice, seo-specialist, content-marketer, competitive-analyst, market-researcher, and commands like /campaign-brief, /blog-post, /email-sequence, /landing-page, /launch-plan.
## Pricing
- [Engineer](https://agentskit.co/#pricing): $49 one-time, lifetime updates.
- [Marketing](https://agentskit.co/#pricing): $49 one-time, lifetime updates.
- [Complete Bundle](https://agentskit.co/#pricing): $89 one-time for the first 50 customers, then $149. Both kits, all 89 agents / 122 skills / 181 commands.
- Every plan: one-time payment, lifetime updates, private-repo delivery, 14-day money-back guarantee. Secure checkout via Polar.
## Key pages
- [Home](https://agentskit.co/): product overview, what's inside, kits, pricing, FAQ.
- [Free tools](https://agentskit.co/tools): free browser tools and Claude skills for people building with Claude Code.
- [Skills & plugins marketplace](https://agentskit.co/marketplace): hand-picked, free, open-source Claude Code skills and plugins, each with a review and GitHub link.
- [Pricing](https://agentskit.co/#pricing): plans and what each includes.
- [FAQ](https://agentskit.co/#faq): what you get, framework lock-in, updates, refunds.
## Free tools
- [SKILL.md Validator](https://agentskit.co/claude-skill-md-validator): paste a Claude Code SKILL.md and instantly lint the YAML frontmatter, name and description rules, body length, and when-to-use structure, with the exact fixes. Runs 100% in your browser.
- [Website Security Audit](https://agentskit.co/claude-skill-for-website-security-audit): a free Claude skill that audits your own website or codebase for missing HTTPS, exposed files, hardcoded secrets, and vulnerable dependencies, then returns a prioritized plain-English fix-it report.
## Free resources (guides & prompt library)
- [Claude prompt library](https://agentskit.co/prompts): 50 field-tested Claude prompts across five collections, each with the reasoning for why it works.
- [Claude prompts for data analysis](https://agentskit.co/prompts/claude-prompts-for-data-analysis): 10 prompts covering dataset profiling, hypothesis-first analysis, SQL with guardrails, statistical sanity checks, executive translation.
- [Claude prompts for writing](https://agentskit.co/prompts/claude-prompts-for-writing): 10 prompts covering voice-matched drafts, ruthless editing, de-AI-ifying text, structure diagnosis, titles.
- [Claude prompts for business](https://agentskit.co/prompts/claude-prompts-for-business): 10 prompts covering strategy red-teaming, pricing decisions, negotiation prep, hiring scorecards, decision journals.
- [Claude prompts for coding](https://agentskit.co/prompts/claude-prompts-for-coding): 10 prompts covering hypothesis-driven debugging, adversarial code review, refactoring plans, bug-hunting tests, security pre-flight.
- [Claude prompts for marketing](https://agentskit.co/prompts/claude-prompts-for-marketing): 10 prompts covering positioning, landing pages, customer-language mining, email sequences, SEO briefs, ad variants.
- [Blog](https://agentskit.co/blog): practical guides for building with Claude.
- [Getting started with Claude Code](https://agentskit.co/blog/getting-started-with-claude-code): install to first shipped feature: CLAUDE.md, permissions, skills, workflow habits.
- [How to build AI agents with Claude](https://agentskit.co/blog/how-to-build-ai-agents-with-claude): the agent loop, tool use, MCP, Agent SDK, and reliability practices, with code.
- [10 prompting techniques for Claude](https://agentskit.co/blog/10-prompting-techniques-for-claude): context front-loading, examples, XML tags, chain-of-thought, role prompts, and more, each with before/after.
- [7 Claude prompts for smarter investing](https://agentskit.co/blog/7-claude-prompts-for-smarter-investing): research workflows (annual reports, thesis red-teaming, decision journals). Educational, not financial advice.
- [Use cases](https://agentskit.co/use-cases): honest, role-by-role maps of where Claude changes real work.
- [Claude for developers](https://agentskit.co/use-cases/claude-for-developers) · [Claude for data scientists](https://agentskit.co/use-cases/claude-for-data-scientists) · [Claude for product managers](https://agentskit.co/use-cases/claude-for-product-managers) · [Claude for students](https://agentskit.co/use-cases/claude-for-students)
- [Comparisons](https://agentskit.co/comparisons): hype-free tool comparisons with bias disclosed.
- [AgentsKit vs Superpowers](https://agentskit.co/comparisons/agentskit-vs-superpowers) · [AgentsKit vs Everything Claude Code](https://agentskit.co/comparisons/agentskit-vs-everything-claude-code) · [AgentsKit vs ClaudeKit (now AgentKit)](https://agentskit.co/comparisons/agentskit-vs-claudekit)
- [Claude vs ChatGPT](https://agentskit.co/comparisons/claude-vs-chatgpt) · [Claude Sonnet vs Opus](https://agentskit.co/comparisons/claude-sonnet-vs-opus) · [Claude vs GitHub Copilot](https://agentskit.co/comparisons/claude-vs-copilot)
- [Claude coding tools guide](https://agentskit.co/tools/claude-coding-tools): the 2026 stack, covering Claude Code, IDE surfaces, MCP connectors, skills layer.
- [Claude MCP servers guide](https://agentskit.co/tools/claude-mcp-servers): what MCP is, the connector categories worth installing, and how to add them safely.
- [Claude productivity tools guide](https://agentskit.co/tools/claude-productivity-tools): Projects, Artifacts, connectors, skills, and power-user habits.
## MCP server (free, no account)
AgentsKit runs a free, unauthenticated MCP server exposing every guide on this site plus the free tools, so an agent can read them directly instead of scraping pages.
- Endpoint: https://agentskit.co/api/mcp (HTTP transport, JSON-RPC 2.0)
- Install: `claude mcp add --transport http agentskit https://agentskit.co/api/mcp`
- Tools: `list_posts`, `get_post`, `search_posts`, `list_free_tools`
## Free skills & plugins marketplace
- [Harvey](https://agentskit.co/marketplace/harvey-ai-sales-agent): open-source autonomous AI sales agent that runs on a Claude Code or Codex subscription; finds prospects, writes cold email, handles replies.
- [/brag](https://agentskit.co/marketplace/brag-launch-video-skill): agent skill that turns the current project into a short launch video with music and share copy.
- [GEO-SEO Claude](https://agentskit.co/marketplace/geo-seo-claude): Claude Code skill that audits a site for AI search (GEO) and classic SEO, with a 0-100 score and PDF report.
- [Hypit](https://agentskit.co/marketplace/hypit-ai-video-cloning): open-source skill that lets Claude Code or Codex clone a viral video as an editable, word-timed workflow and render variants.
- [REA](https://agentskit.co/marketplace/rea-reverse-engineering-mcp): MCP server that lets coding agents reverse engineer native binaries, Electron/JavaScript apps, .NET, Android and websites, with evidence.
- [API Anything](https://agentskit.co/marketplace/api-anything-website-to-api): turns a website into operations an agent calls over MCP as plain HTTP, learned from the site's own frontend and self-repairing.
## Optional
- [Terms of Service](https://agentskit.co/terms)
- [Privacy Policy](https://agentskit.co/privacy)
- [Refund Policy](https://agentskit.co/refund)
- [Legal overview](https://agentskit.co/legal)
- [Disclaimer](https://agentskit.co/disclaimer)
- Contact: epictools.io@gmail.com