Skip to content
This repository was archived by the owner on May 15, 2026. It is now read-only.

Commit 9f77362

Browse files
committed
feat: add AgentBoot + P2PCLAW integration examples
Add example configurations and documentation for bootstrapping research agents using the P2PCLAW decentralized network with Roo Code custom modes and MCP server support. - Example .roomodes entry for AgentBoot custom mode - MCP server configuration snippet for P2PCLAW - Documentation with workflow examples Relates to #12307
1 parent ad25634 commit 9f77362

3 files changed

Lines changed: 146 additions & 0 deletions

File tree

Lines changed: 86 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,86 @@
1+
# AgentBoot + P2PCLAW Integration for Roo Code
2+
3+
This example shows how to use Roo Code's custom modes and MCP server support to bootstrap specialized research agents using [P2PCLAW](https://github.com/Agnuxo1/OpenCLAW-P2P) and [AgentBoot](https://github.com/Agnuxo1/AgentBoot).
4+
5+
## Overview
6+
7+
**AgentBoot** is a bootstrap agent from the P2PCLAW decentralized scientific research network. It creates new specialized AI agents on demand. By combining Roo Code's multi-mode architecture with AgentBoot templates, you can turn a generic Roo Code agent into a bare-metal sysadmin or research agent in minutes.
8+
9+
### What This Integration Provides
10+
11+
- **AgentBoot custom mode**: Pre-configured system prompts for bare-metal hardware detection and OS installation workflows
12+
- **P2PCLAW MCP server config**: Ready-to-use snippet for connecting Roo Code to the P2PCLAW mesh network
13+
- **Workflow templates**: Step-by-step guidance for bootstrapping research agents
14+
15+
## Quick Start
16+
17+
### Step 1: Add the AgentBoot Custom Mode
18+
19+
Copy the contents of [`roomodes-example.yaml`](./roomodes-example.yaml) into your project's `.roomodes` file (create it in your project root if it doesn't exist).
20+
21+
This adds an "AgentBoot" mode to Roo Code with system prompts tailored for:
22+
23+
- Hardware inventory and detection
24+
- OS installation and configuration
25+
- Agent bootstrapping and registration on the P2PCLAW network
26+
- Research workflow automation
27+
28+
### Step 2: Configure the P2PCLAW MCP Server
29+
30+
Add the P2PCLAW MCP server to your Roo Code MCP settings. Copy the contents of [`mcp-config-example.json`](./mcp-config-example.json) into your MCP configuration.
31+
32+
To configure MCP servers in Roo Code:
33+
34+
1. Open the Roo Code sidebar
35+
2. Click the MCP servers icon (plug icon)
36+
3. Add a new server using the configuration from `mcp-config-example.json`
37+
38+
Alternatively, add the server entry to your `~/.roo/mcp.json` or project-level `.roo/mcp.json` file.
39+
40+
### Step 3: Use the Integration
41+
42+
Once configured, switch to the **AgentBoot** mode in Roo Code and start bootstrapping agents.
43+
44+
## Example Workflow
45+
46+
### Creating a Research Agent
47+
48+
```
49+
User: "Create an agent that analyzes protein folding papers"
50+
51+
1. Switch to AgentBoot mode in Roo Code
52+
2. AgentBoot mode guides you through:
53+
a. Defining the agent's research domain (protein folding)
54+
b. Specifying data sources (PubMed, arXiv, bioRxiv)
55+
c. Configuring the agent's analysis capabilities
56+
3. The P2PCLAW MCP server registers the new agent on the network
57+
4. The agent joins the P2PCLAW ecosystem with full Tribunal scoring
58+
```
59+
60+
### Bare-Metal Bootstrapping
61+
62+
```
63+
User: "Set up a new compute node for the research cluster"
64+
65+
1. Switch to AgentBoot mode
66+
2. AgentBoot detects available hardware via system commands
67+
3. Guides OS installation and dependency setup
68+
4. Configures the node for CAJAL (local LLM) support
69+
5. Registers the node on the P2PCLAW mesh network
70+
```
71+
72+
## File Reference
73+
74+
| File | Description |
75+
|------|-------------|
76+
| [`roomodes-example.yaml`](./roomodes-example.yaml) | Example `.roomodes` entry defining the AgentBoot custom mode |
77+
| [`mcp-config-example.json`](./mcp-config-example.json) | MCP server configuration snippet for the P2PCLAW server |
78+
79+
## Links
80+
81+
- [P2PCLAW](https://github.com/Agnuxo1/OpenCLAW-P2P) -- Decentralized scientific research network
82+
- [AgentBoot](https://github.com/Agnuxo1/AgentBoot) -- Bootstrap agent for creating specialized agents
83+
- [P2PCLAW MCP Server](https://github.com/Agnuxo1/p2pclaw-mcp-server) -- MCP server for P2PCLAW integration
84+
- [CAJAL](https://github.com/Agnuxo1/CAJAL) -- Local LLM engine
85+
- [Roo Code Custom Modes](https://docs.roocode.com/advanced-usage/custom-modes) -- How to use custom modes
86+
- [Roo Code MCP Support](https://docs.roocode.com/features/mcp) -- How to configure MCP servers
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
{
2+
"mcpServers": {
3+
"p2pclaw": {
4+
"command": "npx",
5+
"args": ["-y", "@agnuxo/p2pclaw-mcp-server"],
6+
"env": {
7+
"P2PCLAW_NETWORK_ID": "your-network-id",
8+
"P2PCLAW_NODE_NAME": "your-node-name"
9+
},
10+
"disabled": false,
11+
"alwaysAllow": []
12+
}
13+
}
14+
}
Lines changed: 46 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,46 @@
1+
# AgentBoot Custom Mode for Roo Code
2+
#
3+
# Add this entry to your project's .roomodes file to enable the AgentBoot mode.
4+
# If .roomodes doesn't exist yet, create it in your project root with:
5+
#
6+
# customModes:
7+
# - <paste the entry below>
8+
#
9+
# For more information on custom modes, see:
10+
# https://docs.roocode.com/advanced-usage/custom-modes
11+
12+
- slug: agentboot
13+
name: "\U0001F9EA AgentBoot"
14+
roleDefinition: |-
15+
You are AgentBoot, a specialized bootstrap agent for the P2PCLAW decentralized
16+
scientific research network. Your expertise includes:
17+
18+
- Bare-metal hardware detection and inventory (CPU, GPU, RAM, storage, network)
19+
- Operating system installation and configuration for research workloads
20+
- Dependency management for scientific computing environments
21+
- Agent specification and registration on the P2PCLAW mesh network
22+
- Research workflow design and automation
23+
- CAJAL local LLM engine setup and configuration
24+
25+
You guide users through the complete lifecycle of creating a new research agent:
26+
1. Environment assessment -- detect available hardware and software
27+
2. Infrastructure setup -- install and configure OS, drivers, and dependencies
28+
3. Agent design -- define the agent's research domain, capabilities, and tools
29+
4. Network registration -- register the agent on P2PCLAW via MCP
30+
5. Validation -- verify the agent is operational and connected to the mesh
31+
32+
You work methodically, confirming each step before proceeding. You use system
33+
commands to detect hardware, install packages, and verify configurations. When
34+
interacting with P2PCLAW, you use the MCP tools provided by the p2pclaw-mcp-server.
35+
whenToUse: >-
36+
Use this mode when you need to bootstrap a new research agent, set up a
37+
bare-metal compute node, or register an agent on the P2PCLAW decentralized
38+
research network. Ideal for hardware detection, OS installation workflows,
39+
and creating specialized agents for scientific research tasks.
40+
description: Bootstrap research agents and bare-metal nodes for the P2PCLAW network.
41+
groups:
42+
- read
43+
- edit
44+
- command
45+
- mcp
46+
source: project

0 commit comments

Comments
 (0)