

Agent Builder Tool
Model
Available ActionsEach successful request consumes credits as outlined below.
create_newupdate_existingfetch_existingsearch_publicpublisharchiveremixadd_productremove_productadd_workflowremove_workflowattach_contextdetach_contextadd_showcase_exampleremove_showcase_example
Details
Build your own AI agent in a conversation. Describe the job you want done and this AI agent builder creates the agent, gives it the tools and workflows it needs, writes its operating instructions, and puts it live under its own name and link. Start from a blank agent or branch off one that already works, then keep shaping it in plain language: connect a CRM, an inbox, a calendar, a phone line, research and data tools, or an entire multi step workflow, and attach your own context documents so it answers with your pricing, your policies, and your voice instead of generic advice. Preview the agent in a live chat, publish it when it is ready, hand it to your team, and spin up new agents for other roles while the original keeps running untouched. No code, no framework, and no prompt engineering background required. If you can explain the job to a new hire, you can build the AI agent that does it.
Use Cases
Build a custom AI agent without writing code, Turn a job description into a working AI agent, Equip an agent with CRM inbox calendar and messaging tools, Attach multi step workflows to an agent, Ground an agent in your own policies pricing and brand voice, Publish an AI agent and share it with your team, Remix an existing agent into a specialized version, Add showcase chat previews to an agent listing, Manage the agent lifecycle from draft to published to archived, Prototype and tune agent instructions in a live chat
Dynamic MCP Setup
Connect once through AgentPMT Dynamic MCP, then use approved tools from the same agent connection.
30 Second Setup
STDIO connector for Claude Code, Codex, Cursor, Zed, and other LLMs that require STDIO or custom connections.
npm install -g @agentpmt/mcp-routeragentpmt-setupHosted Streamable HTTPS
MCP endpoint for browser-based apps like ChatGPT, Claude, Grok, or any time you want a streamable connection with no local install.
https://api.agentpmt.com/mcpConfig Example
Use the hosted endpoint directly in clients that support remote MCP. Store your Bearer token in the client config or secret field.
{
"mcpServers": {
"agentpmt": {
"type": "streamable-http",
"url": "https://api.agentpmt.com/mcp",
"headers": {
"Authorization": "Bearer <AGENTPMT_BEARER_TOKEN>",
"x-instance-metadata": "{\"client\":\"generic-mcp\",\"platform\":\"remote\"}"
}
}
}
}Need client videos, organization controls, audit details, and the full feature overview?
More About Dynamic MCPAbout this Product
Build your own AI agent, in a conversation
Agent Builder Tool is a no code AI agent builder. You describe the job you want done, and it creates the agent, connects the tools and workflows that job needs, writes the operating instructions, and publishes it under its own name and link. Everything happens in plain language, in chat, while you watch it take shape.
Most AI agent platforms hand you a framework and a blank file. This one hands you a working agent. Tell it what the agent should know, what it should be able to do, and what it must never do, and it assembles a custom AI agent you can run, share, and keep improving.
What you can build
- Sales agents that read your CRM pipeline, draft follow ups in your voice, and log every touch.
- Support agents that answer from your own policy and pricing documents instead of guessing.
- Research agents that monitor news, competitors, filings, and public data on a schedule.
- Operations agents that move work between your inbox, calendar, spreadsheets, storage, and messaging tools.
- Front desk agents that answer the phone, qualify inbound leads, and book the appointment.
- Internal copilots that know how your team actually works and follow your rules every time.
How it works
- Describe the agent. Say what it should do in the same words you would use to brief a new hire. A private draft agent is created immediately.
- Give it tools. Attach any tools from the marketplace: CRM, email, calendar, SMS and voice, documents, spreadsheets, maps, payments, data, and more. Your agent can use every one of them.
- Attach workflows. Add complete multi step workflows so your agent can run an entire process, not just a single call.
- Teach it your business. Attach context documents with your pricing, policies, product details, and brand voice so answers come out right the first time.
- Write the rules. The system prompt is set in plain language and can be rewritten any time, so behavior stays exactly where you want it.
- Test it live. Chat with the agent, see what it does, and adjust on the spot.
- Publish and share. Move the draft to active, give it a public link, and put it in front of your team or your customers.
Real tools, not toy integrations
An AI agent is only as useful as what it can reach. Agent Builder Tool connects your agent to live tools that read and write real systems, and it handles the plumbing for you. When a tool or workflow needs file storage, file handling is attached to the agent automatically, so uploads, exports, and attachments simply work.
Workflows turn agents into operators
Single tool calls answer questions. Workflows finish jobs. Attach a multi step workflow and your agent can run a full pipeline: pull the records, do the analysis, draft the output, check the quality, file it in the right place, and report back. One agent can carry several workflows and choose the right one for the request.
Your knowledge, your voice
Context documents give your agent the things a generic model cannot know: how you price, what you promise, which words you never use, who to escalate to. Attach up to ten per agent and swap them whenever the business changes.
Publish, share, and branch
Publishing turns a draft into a live agent with its own page and link. Remixing creates a new private draft from any agent you can see and credits the original as its source. The remix starts with every tool, workflow, context document, operating instruction, and available model that you can access already copied; anything unavailable to you is skipped and reported without changing the original. Build the sales version, remix it for renewals, remix that for onboarding, then tune each branch for its own job while the proven agent keeps running untouched. Showcase chat previews let visitors see exactly what your agent does before they run it.
Who it is for
Founders, operators, marketers, sales teams, consultants, agencies, and developers who want custom AI agents without a build cycle. If you have been searching for an AI agent builder, an AI agent creator, or a no code way to build AI agents that use real tools, this is the tool that does it, from first idea to published agent, in one conversation.
Frequently Asked Questions
How do I connect this tool to an external agent?
You can install the local MCP server by opening a terminal and running:
Install commands
npm install -g @agentpmt/mcp-router
agentpmt-setupThis will connect you to local agents like Claude Code, Windsurf, Grok Build, Cursor, etc.
Alternatively you can connect to the hosted version with this config block, no installation required:
Hosted MCP config
{
"mcpServers": {
"agentpmt": {
"type": "streamable-http",
"url": "https://api.agentpmt.com/mcp",
"headers": {
"Authorization": "Bearer <AGENTPMT_BEARER_TOKEN>",
"x-instance-metadata": "{\"client\":\"generic-mcp\",\"platform\":\"remote\"}"
}
}
}
}View MCP Connection Instructions for more details.
How does an external agent use this tool?
After the external agent is connected to an Agent Group that can use this tool, paste this prompt into the agent:
Agent prompt
Use the AgentPMT-Tool-Search-and-Execution tool. First call action 'get_instructions' so you know how to use the tool search interface. Then call action 'get_schema' with tool_id 69ed32a83ccc082ef9976212 ("Agent Builder Tool"). After reading the schema and any returned instructions, tell me what this tool can do, we are going to be using it
The agent should fetch the tool schema first, collect the required parameters for your request, and then call the tool through AgentPMT.
Can I build an AI agent without coding?
Yes. The entire build happens in conversation. You describe the agent, review what gets attached, adjust the instructions in plain language, and publish when it behaves the way you want. There is no framework to install, no deployment step, and no prompt engineering background required.
Can I change an agent after it is published?
Yes. Name, description, instructions, tools, workflows, and context documents can all be updated on a live agent, and changes take effect on the next conversation. If an agent has run its course you can archive it and point its old links at a replacement, so nothing sends visitors to a dead end.
Can I share my AI agent with my team or publish it publicly?
Yes. Agents start as private drafts so you can experiment safely. Publishing makes the agent active with its own page and link, ready to hand to your team. You can also add showcase chat previews so anyone opening the agent can see exactly what it does before they run it.
How do I give my AI agent tools like a CRM or email?
Ask for them by name. Tools from the marketplace can be attached to any agent you own, including CRM, email, calendar, SMS and voice, spreadsheets, documents, storage, maps, payments, and data services. Once attached, the agent can use those tools during a chat. If a tool needs file storage, file handling is attached to the agent automatically.
How does my agent learn about my business?
Attach context documents. These hold your pricing, policies, product details, escalation rules, and brand voice, and the agent reads them before it answers. You can attach up to ten documents to an agent and swap them out whenever the business changes, so answers stay current without rebuilding anything.
What does remixing an agent mean?
Remixing creates a new private draft from any agent you can view and credits the original as its source. The remix copies every tool, workflow, Agent Context document, operating instruction, and available model that you can access. Missing, inactive, inaccessible, or unavailable items are skipped and reported so the usable parts still copy successfully. The original agent is never modified, making remix the fastest way to branch a proven agent for renewals, onboarding, or another specialized role.
What is an AI agent builder?
An AI agent builder is a tool that turns a description of a job into a working AI agent. Instead of writing code, you explain what the agent should do, what it should have access to, and how it should behave. Agent Builder Tool creates the agent, attaches the tools and workflows it needs, sets its operating instructions, and publishes it with its own name and link.
What is the difference between a tool and a workflow?
A tool is a single capability, such as sending an email or updating a deal. A workflow is a complete multi step process that chains several tools together with logic and checks along the way. Attaching a workflow lets your agent finish an entire job in one request instead of improvising the steps. An agent can carry several workflows and pick the right one for the situation.
