infra-networking

Identify and describe AICP network surfaces including port mappings and MCP transport.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/cyberpunk042/devops-expert-local-ai --skill infra-networking
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: infra-networking
Source: https://github.com/cyberpunk042/devops-expert-local-ai/tree/main/.claude/skills/infra-networking
Command: npx skills add https://github.com/cyberpunk042/devops-expert-local-ai --skill infra-networking

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manage AICP's network surfaces — port mappings, MCP transport, planned LocalAI P2P cluster peering, and reverse-proxy/firewall integration. AICP is local-first by default (everything on localhost); fleet rollout extends networking to LAN with careful controls.

Core Features & Use Cases

  • Inventory ports and surfaces: document host-to-container mappings and metrics endpoints.
  • Expose LocalAI for fleet access: provide guidance to safely open selected ports to LAN with firewall considerations.
  • Set up LocalAI P2P cluster peering (planned): outline topology prerequisites and documentation needs.
  • Place AICP behind a reverse proxy: configure TLS and proxy routing with authentication considerations.
  • Diagnose MCP connection failures: verify stdio transport configuration and PATH/cmd invocation issues.

Quick Start

Review the firewall rules and port mappings in docker-compose.yaml and perform a pre-deployment network audit.

Frequently Asked Questions about infra-networking

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit LocalAI port mappings before deploying AICP?

To audit LocalAI port mappings, review docker-compose.yaml to verify host 8090 maps to container 8080, metrics run on 9101, and optional Prometheus and Grafana run on 9090 and 3000. Ensure localhost exposure defaults are enforced before deployment.

How do I configure firewall rules to expose LocalAI ports to a LAN fleet?

To configure firewall rules for LAN fleet access, carefully open selected LocalAI ports beyond the default localhost exposure. Operator-controlled port openings must be documented in operator runbooks to maintain secure fleet rollout and network surface management.

Why does my MCP stdio transport connection fail when invoking the agent server?

MCP stdio transport connection failures often stem from incorrect PATH environment variables or cmd invocation issues. Verify the stdio transport configuration and ensure the agent server port is correctly mapped and accessible for MCP connectivity troubleshooting.

Can I place AICP behind a reverse proxy with TLS and authentication?

Yes, you can place AICP behind a reverse proxy to configure TLS and proxy routing with authentication considerations. This approach helps manage AICP network surfaces and secure LocalAI endpoints when extending infrastructure beyond localhost.

What are the prerequisites for setting up LocalAI P2P cluster peering?

Setting up LocalAI P2P cluster peering requires outlining topology prerequisites and documentation needs. As a planned feature, it extends AICP networking from localhost to LAN, requiring careful firewall configurations and operator-controlled port openings for secure peering.

What limitations exist when extending AICP networking beyond localhost?

Extending AICP networking beyond localhost requires careful firewall rules and operator-controlled port openings, as the system is local-first by default. Fleet rollout to LAN demands strict documentation in operator runbooks to prevent unintended exposure of network surfaces.