# AI Agent > AI Agent is a no-code platform for building, deploying, and running AI agents that automate multi-step workflows across business tools. Teams connect their existing apps — CRM, inbox, analytics, ad platforms, help desks, and more — configure what an agent should do using plain-language instructions, set approval gates for actions that need human sign-off, and run durable workflows that handle research, drafting, triage, and reporting without manual step-by-step prompting. The platform is designed so that the people closest to a process — not just engineers — can build and adjust agents directly. Agents work through a visual builder, connect to real integrations, and can run on a schedule, on an event trigger, or on demand. Every run is logged with its inputs, steps, and outcomes for auditing and iteration. ## Core pages - [Home](https://aiagent.app): The homepage introduces AI Agent as an AI automation platform for every team and business. It covers the core value proposition — build, deploy, and run AI agents that automate busywork including research, workflows, reports, and more — and links to use cases, pricing, and the agent builder. The positioning is generalist: agents for operations, sales, marketing, support, and any team with repetitive multi-tool workflows. - [Pricing](https://aiagent.app/pricing): The pricing page details subscription plans from Free through Max, with a feature comparison table covering agents per month, tasks per agent per month, available AI models, integration limits, and support tiers. A free plan is available without a credit card. Paid plans unlock higher credit limits, more integrations, latest AI models, and priority support. An enterprise tier is available for custom volume. - [Product tour](https://aiagent.app/tour): The real dashboard on a sample workspace, read only. Board, inbox, overview, chat, agents and workflows, each with a short guided tour. - [Compare](https://aiagent.app/compare): A side-by-side comparison of AI Agent against alternative AI agent and automation platforms, covering interface type, setup complexity, integration breadth, workflow durability, approval gate support, and pricing. Aimed at buyers evaluating AI Agent against tools like Zapier, Make, n8n, Gumloop, AutoGPT, and similar platforms. - [Templates](https://aiagent.app/templates): 72 ready-made AI agents across sales, marketing, product, operations, people, engineering and personal use. Each template has a public page at https://aiagent.app/templates/ with its full instructions, memories, skills, routines (with schedules) and integrations, credited to its original creator with a source link. Installing one requires an AI Agent workspace; routines install paused. - [Done For You](https://aiagent.app/done-for-you): A managed build service for teams that want custom AI agents designed and deployed for them rather than self-serving through the platform. Covers what can be built, how the engagement works, and how to book a call to discuss a custom plan. - [Content](https://aiagent.app/content): A hub page linking to use cases, blog articles, and AI agent insights. Acts as a directory for teams exploring what AI agents can do before committing to building one. ## Use cases - [Use Cases](https://aiagent.app/usecases): The main use cases index page, listing AI agent workflows across industries, business functions, and specific workflow types. Links out to both curated keyword landing pages (sales, support, SEO, marketing, etc.) and a larger set of programmatic industry and function pages covering hundreds of specific use cases from healthcare and finance to logistics and retail. - [AI Sales Agent](https://aiagent.app/usecases/ai-sales-agent): An AI sales agent connects to a CRM and inbox to research target accounts, find the right contacts, draft personalized outreach and follow-up messages, triage inbound replies, and surface the next best action for every deal. The agent proposes ranked actions and drafts for rep review — nothing is sent without human approval. It also handles pipeline hygiene: flagging stale deals, logging activity, and assembling pre-meeting briefs from connected data sources. - [AI Support Agent](https://aiagent.app/usecases/ai-support-agent): An AI support agent connects to a help desk and knowledge base to triage incoming tickets, classify and tag them by priority and sentiment, and draft grounded replies sourced from the team's own help documentation rather than generated from scratch. For simple, well-understood questions it can resolve directly under rules set by the team; complex cases are escalated to the right queue with full context attached. Teams control how much autonomy to grant per queue. - [AI Marketing Agent](https://aiagent.app/usecases/ai-marketing-agent): An AI marketing agent connects to analytics, CRM, and ad platforms to handle the repetitive, data-heavy assembly work across a marketing operation. It researches audiences and competitors, drafts email copy, ad copy, and content briefs, generates campaign performance summaries on a set cadence, and proposes ranked next actions. Every piece of copy or campaign recommendation goes through human review before anything ships. - [AI SEO Agent](https://aiagent.app/usecases/ai-seo-agent): An AI SEO agent connects to analytics and search data sources to run keyword and competitor research, audit pages against on-page and technical best practices, draft structured content briefs for target keywords, and deliver a ranked, prioritized list of recommendations. All outputs are proposals — the agent does not publish changes; a human reviews and approves what to act on. - [AI Workflow Automation](https://aiagent.app/usecases/ai-workflow-automation): AI workflow automation on this platform goes beyond rule-based if-this-then-that tools. An AI agent can reason over inputs within each workflow step — reading documents, weighing options, writing replies — and chain steps across multiple tools without a human driving each one. Workflows can be triggered by schedule, event, or on demand, and include approval gates at any step where a decision carries real-world consequences. - [No-Code AI Agents](https://aiagent.app/usecases/no-code-ai-agents): No-code AI agents are configured through a visual interface rather than code. Teams authorize integrations, write agent instructions in plain language, assemble multi-step workflows from blocks, and set which actions need human sign-off — all without engineering involvement. Changes to instructions, integrations, or approval gates are config edits, not code deployments, so teams can iterate the same day. - [AI Agent Platform](https://aiagent.app/usecases/ai-agent-platform): An AI agent platform is the infrastructure layer that lets teams build, deploy, and manage agents without standing up custom tooling. This page covers what makes a capable platform — real integration depth, workflow durability, observable run logs, and approval gates as a first-class feature — and who it serves: operations, sales, marketing, support, and product teams with repetitive multi-tool workflows, as well as technical teams who want pre-built orchestration and observability rather than building it themselves. - [Best AI Agents](https://aiagent.app/usecases/best-ai-agents): An evaluation guide for teams assessing AI agent platforms. Covers what to look for: real integration depth (not generic webhooks), workflow durability and observability (every run logged with its full decision trace), approval gates as a first-class feature (not bolted on), and no-code accessibility (a non-engineer can build and modify agents without an engineering queue). Includes a practical framework for testing platforms before committing. - [AI Agent Examples](https://aiagent.app/usecases/ai-agent-examples): Free AI agent finder, no signup. Describe a task or pick a team (sales, marketing, product, operations, people, engineering) and get matching agent examples, such as sales call follow-ups, competitor watching, inbound lead routing, support reply drafting, candidate sourcing, and monthly website audits. Each example shows what the agent does, a suggested schedule, and the apps it uses, and opens the matching template in an AI Agent workspace. - [Autonomous AI Agent](https://aiagent.app/usecases/autonomous-ai-agent): An autonomous AI agent can carry out multi-step work without a human driving each step — planning the steps to reach a goal, executing them in sequence across tools, and handling intermediate decisions. Autonomy is a dial, not a switch: teams start with high oversight and progressively remove approval gates on steps where the agent has proven reliable. Every run is logged with its full decision trace so autonomy can be tuned with real evidence. High-stakes steps — those that change real-world state — pause and propose ranked options for human sign-off before continuing. ## Developer resources (AI Agent) - [Developers](https://aiagent.app/developers): The developer resources hub for AI Agent — a crawlable HTML page listing the MCP servers, OpenAPI, auth.md, llms.txt indexes, API and AI catalogs, agent card, pricing.md, docs, and GitHub org links. Intended for agents and integrators discovering the platform, and for people connecting an AI client to their workspaces. - [developers.md](https://aiagent.app/developers.md): Markdown twin of the developer resources page with the same link inventory. - [developers/llms.txt](https://aiagent.app/developers/llms.txt): Scoped plain-text index of developer URLs on aiagent.app. - [OpenAPI](https://aiagent.app/openapi.json): OpenAPI specification for the public discovery and health endpoints. HTTP stays discovery-only: there is no REST or GraphQL tenant API. Agents work in workspaces through the authenticated MCP server. - [auth.md](https://aiagent.app/auth.md): WorkOS-shaped authentication guide — Google OAuth and session cookies for humans; OAuth 2.1 for MCP clients (authorization code with PKCE, Client ID Metadata Documents preferred, Dynamic Client Registration as fallback, refresh tokens); scopes workspace:read, workspace:write, and workspace:run; revocation under Settings → Connected apps. No API keys, personal access tokens, or machine-to-machine grants, and no register_uri. - [MCP](https://aiagent.app/mcp): Authenticated remote MCP server (Streamable HTTP, POST). An MCP client such as Claude or ChatGPT signs the user in with OAuth 2.1, the user approves scopes on a consent screen, and the client then acts as that user, only in workspaces the user belongs to. It covers workspaces, tickets, agents (ask an agent and get its reply), workflows and autopilots (start, check, and cancel runs), tables, knowledge, projects, and the inbox (read, triage, and reply drafts that wait for approval). Approvals are read-only: approving a pending action stays a human action in the dashboard Inbox. Running agents and workflows spends workspace credits. Protected resource metadata: https://aiagent.app/.well-known/oauth-protected-resource/mcp. Server card: https://aiagent.app/.well-known/mcp/server-card.json. - [Connect Claude or ChatGPT](https://docs.aiagent.app/guides/connect-claude-or-chatgpt/): Step-by-step guide to adding the MCP server as a custom connector in Claude, Claude Code, or ChatGPT, approving access, and revoking it. - [Public MCP](https://aiagent.app/mcp/public): Read-only discovery MCP with no sign-in. Tools: get_product, get_pricing, get_auth_info, list_discovery_urls. - [API catalog](https://aiagent.app/.well-known/api-catalog): Linkset catalog pointing at service documentation including docs and llms.txt. - [AI catalog](https://aiagent.app/.well-known/ai-catalog.json): AI-facing JSON catalog of services. - [Agent card](https://aiagent.app/.well-known/agent-card.json): Agent platform card for automated discovery. - [Agent skills](https://aiagent.app/.well-known/agent-skills/index.json): Agent Skills discovery index. - [llms.md](https://aiagent.app/llms.md): Markdown twin serving the same corpus as this file (llms-full.txt). - [index.md](https://aiagent.app/index.md): Same full corpus as llms-full.txt, served as markdown. - [pricing.md](https://aiagent.app/pricing.md): Markdown pricing tiers — Free $0, Start $49, Pro $149, Max $399, Enterprise custom. - [GitHub org](https://github.com/aiagentapp): Public GitHub organization. The product repository github.com/aiagentapp/aiagent is private; use docs and llms.txt for agent platform configs. - [Docs llms.txt redirect](https://aiagent.app/docs/llms.txt): Redirects to https://docs.aiagent.app/llms.txt. ## Resources - [Docs](https://docs.aiagent.app): Customer-facing product documentation covering getting started, agents, workflows, connections, knowledge, and the changelog. A machine-readable index is at https://docs.aiagent.app/llms.txt and a full corpus at https://docs.aiagent.app/llms-full.txt. - [Blog](https://aiagent.app/blog): Articles covering AI agents, workflow automation, product updates, and the broader AI landscape. Topics include introductions to AI agents, use case deep dives, comparisons of AI models and tools, and automation best practices for business teams. - [What is an AI agent?](https://aiagent.app/blog/what-are-ai-agents): The site's main explainer. Covers how an AI agent works (goal, plan, tools, memory, feedback), AI agents versus chatbots and workflow automation, classic and practical agent types, examples by team with links to live templates, how to build or run one, and guardrails. - [AEO checker](https://aiagent.app/tools/aeo-checker): Free, no-signup tool that checks one URL for AI-search readiness: which AI crawlers robots.txt allows, llms.txt presence, sitemap, JSON-LD types, FAQ, headings, title, description and canonical. It returns a 0-100 score, a fix list and a draft llms.txt. No AI model is called. - [Changelog](https://aiagent.app/changelog): A chronological record of product releases, new features, and improvements to the AI Agent platform. Updated regularly as the platform ships new capabilities. ## About - [About](https://aiagent.app/about): Company page covering why AI Agent exists, the problem it solves, and how the product is built for people and businesses. - [Contact](https://aiagent.app/contact): Contact the AI Agent team at hello@aiagent.app for support, partnership, or general inquiries. - [Terms of Use](https://aiagent.app/legal/terms-of-use): The legal terms governing use of the AI Agent platform, including user responsibilities, acceptable use, and service limitations. - [Privacy Notice](https://aiagent.app/legal/privacy-notice): Describes how AI Agent collects, uses, stores, and protects user data, and what rights users have over their personal information. - [Cookie Policy](https://aiagent.app/legal/cookie): Explains what cookies AI Agent uses, why, and how users can manage cookie consent preferences.