Remote Deployment via SSH

Relevant source files

The following files were used as context for generating this wiki page:

• examples/ssh_deployment.py
• src/ssh/manager.py
• AGENT.md
• SKILLS.md
• src/cli.py
• config/default.json
• README.md

Remote Deployment via SSH in the WIREDUP project is a core capability managed by the SSHManager. This system enables AI agents and developers to securely interact with remote infrastructure by executing commands, managing file transfers, and orchestrating deployment workflows. It is designed to be production-ready, supporting secure key management, connection pooling, and multi-agent coordination.

The system integrates directly with the project's Auto-Wiring Engine to provide agents with seamless access to remote execution capabilities. By utilizing standardized protocols and secure credential handling, the SSH module ensures that remote operations are both efficient and protected.
Sources: README.md:12, src/ssh/manager.py:1-50, AGENT.md:15-20

SSH Management Architecture

The SSH architecture is built around the SSHManager class and the SSHCredentials data structure. The manager handles the lifecycle of remote connections, including establishment, command execution, and cleanup.

Core Components

Component	Description
SSHManager	Primary controller for managing remote connections and executing commands.
SSHCredentials	Data structure containing host, port, username, password, or key file details.
SSHCommandSkill	An agent-facing skill that wraps the manager for use in AI workflows.
RemoteAgent	A specialized agent type designed for deployment on remote servers.

Sources: src/ssh/manager.py (inferred from SSHManager implementation), examples/ssh_deployment.py:24-34, SKILLS.md:88-100, AGENT.md:275-295

Connection Flow

The following diagram illustrates the sequence of operations when executing a remote command via the SSH module.

This flow ensures that every interaction is encapsulated within a managed connection state, providing clear success or failure feedback through the SSHResult structure.
Sources: src/ssh/manager.py:10-40, examples/ssh_deployment.py:38-48

Command Execution and Batching

The system supports both single command execution and batch processing. Batch processing is particularly useful for deployment scripts where multiple sequential steps (e.g., pwd, git pull, make) are required.

Execution Methods

• execute(command, credentials): Establishes a connection, runs a single string command, and returns the output.
• execute_batch(commands, credentials): Takes a list of strings and executes them sequentially over a managed connection.
• ssh_exec (CLI): A command-line interface implementation allowing users to trigger remote commands directly from the terminal.

Sources: src/ssh/manager.py:20-50, examples/ssh_deployment.py:53-65, src/cli.py:204-232

# Batch execution example from source
commands = [
    'pwd',
    'whoami',
    'date',
    'uname -a'
]
results = manager.execute_batch(commands, credentials=credentials)

Sources: examples/ssh_deployment.py:53-61

Remote Agent Deployment

Advanced deployment is handled by the RemoteAgent, which utilizes the SSHManager to bootstrap itself onto a remote server. This process involves uploading the agent's source code and initiating its execution environment.

Deployment Workflow

1. Code Upload: The agent uses upload_file to transfer the implementation script to the target /opt/agents/ directory.
2. Environment Setup: Commands are executed to ensure Python and dependencies are available.
3. Start Agent: A final command triggers the remote agent process.

Configuration and Security

Remote deployment settings are managed through the global configuration system, allowing for secure defaults and environment-specific overrides.

SSH Configuration Options

Option	Type	Default	Description
ssh.enabled	Boolean	false	Global toggle for SSH capabilities.
ssh.default_timeout	Integer	30	Timeout in seconds for connection attempts.
ssh.max_connections	Integer	10	Limit for the connection pool.
ssh.connection_pool.max_age	Integer	3600	Maximum age of a pooled connection in seconds.

Sources: config/default.json:20-30, src/cli.py:52-55

Security Best Practices

• Key Rotation: The system supports SSH_KEY_PATH configuration for using ED25519 or RSA keys instead of passwords.
• Credential Masking: Sensitive data is intended to be handled via environment variables like AI_SSH_KEY_PATH.
• Sandboxing: Remote execution should be restricted to allowed_commands when using the ExecutionAgent subtype to prevent arbitrary system access.

Sources: README.md:15, AGENT.md:150-160, config/default.json:44-50

Summary

Remote Deployment via SSH provides the foundational infrastructure for WIREDUP's distributed agent capabilities. By combining a robust SSHManager with specialized agent types and skills, the project enables automated, secure, and scalable remote environment management. Whether triggered via the CLI, an automated script, or an AI agent, the SSH module ensures consistent execution and detailed reporting through its structured result types and connection management.
Sources: src/ssh/manager.py:1-100, AGENT.md:275-300, README.md:40-50