commit 029eb7e7392d092249a0567173499bab62877570 Author: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Date: Mon Sep 8 11:56:21 2025 +0000 Deployed 1ade0a8 with MkDocs version: 1.6.1 diff --git a/.nojekyll b/.nojekyll new file mode 100644 index 00000000..e69de29b diff --git a/404.html b/404.html new file mode 100644 index 00000000..ee4c6568 --- /dev/null +++ b/404.html @@ -0,0 +1,688 @@ + + + +
+ + + + + + + + + + + + + + +In Rowboat, you can add tools to your agents by (a) selecting from a in-built library of MCP tools (b) adding your own customer MCP servers (c) integrating your APIs through a webhook (e) mocking tool calls to test the system.
+Rowboat has partnered with to provide a growing library of hosted MCP servers. You can obtain a Klavis API key and set it using:
export KLAVIS_API_KEY=your-klavis-api-key
+
+
Enable any of the hosted MCP servers by clicking on the enable button. The server will take approximately 10 seconds to spin up.
+
For most servers, you will need to authorize it by clicking on the 'Auth' button and connecting to your account e.g. connecting to you github or slack account
+The servers you have enabled will show up under tools section in the build view and can be added to any of the agents.
+
Note: For GSuite tools, you need to get a Google Client ID and set it using:
+export KLAVIS_GOOGLE_CLIENT_ID=your-google-client-id
+
+To obtain a Google Client ID: +1. Go to the Google Cloud Console +2. Create a new project or select an existing one +3. Go to "APIs & Services" → "OAuth consent screen" and complete the setup +4. Go to "APIs & Services" → "Credentials" → "OAuth client ID" and create credentials for web application.
+You can add any running MCP server in Settings -> Tools.
+
You can use supergateway to expose any MCP stdio server as an SSE server.
+Now, you can import the tools from the MCP server in the Build view.
+
You can mock any tool you have created by checking the 'Mock tool responses' option.
+
When agents call tools during a chat in the playground, the tool call parameters and response are available for debugging real-time. For testing purposes, the platform can produce mock tool responses in the playground, without integrating actual tools.
+
The description conveys the agent's role in the multi-agent system. Writing a good description is important for other agents to know when to pass control of the conversation to an agent.
+Agent instructions are the backbone of an agent, defining its behavior. RowBoat Studio's copilot produces a good framework for agent instructions, involving Role, Steps to Follow, Scope and Guidelines. Since agents are powered by LLMs, general best practices while writing prompts apply.
+The agent uses examples as a reference for behavior in different scenarios. While there are no prescribed formats to provide examples in, examples should include what the user might say, what the agent should respond with as well as indications of any tool calls to be made.
+Prompts attached to an agent will be used by the agent in addition to instructions.
+Tools attached to an agent will be put out as tool calls. The behavior of when to invoke tools can be fine-tuned by specifying corresponding instructions or prompts. Adding examples to agents can also be useful in controlling tool call behavior.
+In the agent instructions, the connected agents are shown with an '@mention'. If the agent mentioned in an instruction (connected agent) does not actually exist, the connected agent's name would show up with an '!' to call to attention.
+RowBoat currently supports OpenAI LLMs. Agents can be configured to use GPT-4o or GPT-4o-mini.
+ + + + + + + + + + + + + +
+
+
+## Obtain API key and Project ID
+
+Generate API keys via the developer configs in your project. Copy the Project ID from the same page.
+
+
+
+
+## API Endpoint
+
+```
+POST