VMware AIops

English | 中文

AI-powered VMware vCenter/ESXi VM lifecycle and deployment tool — 31 tools across 6 categories.

Companion skills handle everything else:

Skill Scope Install

vmware-monitor Read-only: inventory, health, alarms, events, metrics uv tool install vmware-monitor

vmware-storage Datastores, iSCSI, vSAN management uv tool install vmware-storage

vmware-vks Tanzu Namespaces, TKC cluster lifecycle uv tool install vmware-vks

Need read-only monitoring only? Use VMware-Monitor — zero destructive code in the codebase.

Quick Install (Recommended)

Works with Claude Code, Cursor, Codex, Gemini CLI, Trae, and 30+ AI agents:

# Via Skills.sh
npx skills add zw008/VMware-AIops

# Via ClawHub
clawhub install vmware-aiops

PyPI Install (No GitHub Access Required)

# Install via uv (recommended)
uv tool install vmware-aiops

# Or via pip
pip install vmware-aiops

# China mainland mirror (faster)
pip install vmware-aiops -i https://pypi.tuna.tsinghua.edu.cn/simple

Claude Code Plugin Install

# Add marketplace
/plugin marketplace add zw008/VMware-AIops

# Install plugin
/plugin install vmware-ops

# Use the skill
/vmware-ops:vmware-aiops

Capabilities Overview

What This Skill Does

Category	Tools	Count
VM Lifecycle	power on/off, TTL auto-delete, clean slate	6
Deployment	OVA, template, linked clone, batch clone/deploy	8
Guest Ops	exec commands, upload/download files, provision	5
Plan/Apply	multi-step planning with rollback	4
Cluster	create, delete, HA/DRS config, add/remove hosts	6
Datastore	browse files, scan for images	2

CLI vs MCP: Which Mode to Use

Scenario	Recommended	Why
Local/small models (Ollama, Qwen <32B)	CLI	~2K tokens context vs ~10K for MCP; small models struggle with many tool schemas
Token-sensitive workflows	CLI	SKILL.md + Bash tool = minimal overhead
Cloud models (Claude, GPT-4o)	Either	Both work; MCP gives structured JSON I/O
Automated pipelines / Agent chaining	MCP	Type-safe parameters, structured output, no shell parsing
Monitoring / storage / K8s	Companion skills	See vmware-monitor, vmware-storage, vmware-vks

Rule of thumb: Use CLI for cost efficiency and small models. Use MCP for structured automation with large models.

Architecture

User (Natural Language)
  ↓
AI CLI Tool (Claude Code / Gemini / Codex / Aider / Continue / Trae / Kimi)
  ↓ reads SKILL.md / AGENTS.md / rules
  ↓
vmware-aiops CLI
  ↓ pyVmomi (vSphere SOAP API)
  ↓
vCenter Server ──→ ESXi Cluster ──→ VM
    or
ESXi Standalone Host ──→ VM

Version Compatibility

vSphere Version	Support	Notes
8.0 / 8.0U1-U3	✅ Full	`CreateSnapshot_Task` deprecated → use `CreateSnapshotEx_Task`
7.0 / 7.0U1-U3	✅ Full	All APIs supported
6.7	✅ Compatible	Backward-compatible, tested
6.5	✅ Compatible	Backward-compatible, tested

pyVmomi auto-negotiates the API version during SOAP handshake — no manual configuration needed. The same codebase manages both 7.0 and 8.0 environments seamlessly.

Common Workflows

Deploy a Lab Environment

Browse datastore for OVA images → vmware-aiops datastore browse <ds> --pattern "*.ova"
Deploy VM from OVA → vmware-aiops deploy ova ./image.ova --name lab-vm --datastore ds1
Install software inside VM → vmware-aiops vm guest-exec lab-vm --cmd /bin/bash --args "-c 'apt-get install -y nginx'" --user root
Create baseline snapshot → vmware-aiops vm snapshot-create lab-vm --name baseline
Set TTL for auto-cleanup → vmware-aiops vm set-ttl lab-vm --minutes 480

Batch Clone for Testing

Create plan: vm_create_plan with multiple clone + reconfigure steps
Review plan with user (shows affected VMs, irreversible warnings)
Apply: vm_apply_plan executes sequentially, stops on failure
If failed: vm_rollback_plan reverses executed steps
Set TTL on all clones for auto-cleanup

Migrate VM to Another Host

Check VM info via vmware-monitor → verify power state and current host
Migrate: vmware-aiops vm migrate my-vm --to-host esxi-02
Verify migration completed

VM Lifecycle

Operation	Command	Confirmation	vCenter	ESXi
Power On	`vm power-on <name>`	—	✅	✅
Graceful Shutdown	`vm power-off <name>`	Double	✅	✅
Force Power Off	`vm power-off <name> --force`	Double	✅	✅
Reset	`vm reset <name>`	—	✅	✅
Suspend	`vm suspend <name>`	—	✅	✅
Create VM	`vm create <name> --cpu --memory --disk`	—	✅	✅
Delete VM	`vm delete <name>`	Double	✅	✅
Reconfigure	`vm reconfigure <name> --cpu --memory`	Double	✅	✅
Create Snapshot	`vm snapshot-create <name> --name <snap>`	—	✅	✅
List Snapshots	`vm snapshot-list <name>`	—	✅	✅
Revert Snapshot	`vm snapshot-revert <name> --name <snap>`	—	✅	✅
Delete Snapshot	`vm snapshot-delete <name> --name <snap>`	—	✅	✅
Clone VM	`vm clone <name> --new-name <new>`	—	✅	✅
vMotion	`vm migrate <name> --to-host <host>`	—	✅	❌
Set TTL	`vm set-ttl <name> --minutes <n>`	—	✅	✅
Cancel TTL	`vm cancel-ttl <name>`	—	✅	✅
List TTLs	`vm list-ttl`	—	✅	✅
Clean Slate	`vm clean-slate <name> [--snapshot baseline]`	Double	✅	✅
Guest Exec	`vm guest-exec <name> --cmd /bin/bash --args "..."`	—	✅	✅
Guest Exec (with output)	`vm guest-exec-output <name> --cmd "df -h"`	—	✅	✅
Guest Upload	`vm guest-upload <name> --local f.sh --guest /tmp/f.sh`	—	✅	✅
Guest Download	`vm guest-download <name> --guest /var/log/syslog --local ./syslog`	—	✅	✅

Guest Operations require VMware Tools running inside the guest OS. guest-exec-output auto-detects Linux/Windows shell and captures stdout/stderr.

Plan → Apply (Multi-step Operations)

For complex operations involving 2+ steps or 2+ VMs, use the plan/apply workflow instead of executing individually:

Step	What Happens
1. Create Plan	AI calls `vm_create_plan` — validates actions, checks targets in vSphere, generates plan with rollback info
2. Review	AI shows plan to user: steps, affected VMs, irreversible warnings
3. Apply	`vm_apply_plan` executes sequentially; stops on failure
4. Rollback (if failed)	Asks user whether to rollback, then `vm_rollback_plan` reverses executed steps (irreversible steps skipped)

Plans stored in ~/.vmware-aiops/plans/, auto-deleted on success, auto-cleaned after 24h.

VM Deployment & Provisioning

Operation	Command	Speed	vCenter	ESXi
Deploy from OVA	`deploy ova <path> --name <vm>`	Minutes	✅	✅
Deploy from Template	`deploy template <tmpl> --name <vm>`	Minutes	✅	✅
Linked Clone	`deploy linked-clone --source <vm> --snapshot <snap> --name <new>`	Seconds	✅	✅
Attach ISO	`deploy iso <vm> --iso "[ds] path/to.iso"`	Instant	✅	✅
Convert to Template	`deploy mark-template <vm>`	Instant	✅	✅
Batch Clone	`deploy batch-clone --source <vm> --count <n>`	Minutes	✅	✅
Batch Deploy (YAML)	`deploy batch spec.yaml`	Auto	✅	✅

Cluster Management

Operation	Command	Confirmation	vCenter	ESXi
Cluster Info	`cluster info <name>`	—	✅	❌
Create Cluster	`cluster create <name> [--ha] [--drs]`	—	✅	❌
Delete Cluster	`cluster delete <name>`	Double	✅	❌
Add Host	`cluster add-host <cluster> --host <host>`	Double	✅	❌
Remove Host	`cluster remove-host <cluster> --host <host>`	Double	✅	❌
Configure HA/DRS	`cluster configure <name> [--ha/--no-ha] [--drs/--no-drs]`	Double	✅	❌

Datastore Browser

Feature	vCenter	ESXi	Details
Browse Files	✅	✅	List files/folders in any datastore path
Scan Images	✅	✅	Discover ISO, OVA, OVF, VMDK across all datastores

Scheduled Scanning & Notifications

Feature	Details
Daemon	APScheduler-based, configurable interval (default 15 min)
Multi-target Scan	Sequentially scan all configured vCenter/ESXi targets
Scan Content	Alarms + Events + Host logs (hostd, vmkernel, vpxd)
Log Analysis	Regex pattern matching: error, fail, critical, panic, timeout, corrupt
Structured Log	JSONL output to `~/.vmware-aiops/scan.log`
Webhook	Slack, Discord, or any HTTP endpoint
Daemon Management	`daemon start/stop/status`, PID file, graceful shutdown

Safety Features

Feature	Details
Dry-Run Mode	`--dry-run` on any destructive command prints exact API calls without executing
Plan → Confirm → Execute → Log	Structured workflow: show current state, confirm changes, execute, audit log
Double Confirmation	All destructive ops (power-off, delete, reconfigure, snapshot-revert/delete, clone, migrate) require 2 sequential confirmations — no bypass flags
Rejection Logging	Declined confirmations are recorded in the audit trail
Audit Trail	All operations logged to `~/.vmware-aiops/audit.log` (JSONL) with before/after state
Input Validation	VM name, CPU (1-128), memory (128-1048576 MB), disk (1-65536 GB) validated
Password Protection	`.env` file loading with permission check; never in shell history
SSL Self-signed Support	`disableSslCertValidation` — only for ESXi with self-signed certs in isolated labs; production should use CA-signed certificates
Prompt Injection Protection	vSphere event messages and host logs are truncated, stripped of control characters, and wrapped in boundary markers before output
Webhook Data Scope	Sends notifications to user-configured URLs only — no third-party services by default
Task Waiting	All async operations wait for completion and report result
State Validation	Pre-operation checks (VM exists, power state correct)

vCenter vs ESXi Comparison

Capability	vCenter	ESXi Standalone
vMotion migration	✅	❌
Cross-host clone	✅	❌
Cluster management	✅	❌
All VM lifecycle ops	✅	✅
OVA/Template/Linked Clone deploy	✅	✅
Datastore browsing & image scan	✅	✅
Snapshots	✅	✅
Guest operations	✅	✅

Inventory, alarms, events, sensors, host services, and scanning are now in vmware-monitor.

Troubleshooting

"VM not found" error

VM names are case-sensitive in vSphere. Use exact name from vmware-monitor inventory vms.

Guest exec returns empty output

Use vm_guest_exec_output instead of vm_guest_exec — it auto-captures stdout/stderr. Basic vm_guest_exec only returns exit code.

Deploy OVA times out

Large OVA files (>10GB) may exceed the default 120s timeout. The upload happens via HTTP NFC lease — ensure network between the machine running vmware-aiops and ESXi is stable.

Plan apply fails mid-way

Run vmware-aiops plan list to see failed plan status. Ask user if they want to rollback with vm_rollback_plan. Irreversible steps (delete_vm) are skipped during rollback.

Connection refused / SSL error

Verify target is reachable: vmware-aiops doctor
For self-signed certs: set disableSslCertValidation: true in config.yaml (lab environments only)

Supported AI Platforms

Platform	Status	Config File	AI Model
Claude Code	✅ Native Skill	`skills/vmware-aiops/SKILL.md`	Anthropic Claude
Gemini CLI	✅ Extension	`gemini-extension/GEMINI.md`	Google Gemini
OpenAI Codex CLI	✅ Skill + AGENTS.md	`codex-skill/AGENTS.md`	OpenAI GPT
Aider	✅ Conventions	`codex-skill/AGENTS.md`	Any (cloud + local)
Continue CLI	✅ Rules	`codex-skill/AGENTS.md`	Any (cloud + local)
Trae IDE	✅ Rules	`trae-rules/project_rules.md`	Claude/DeepSeek/GPT-4o/Doubao
Kimi Code CLI	✅ Skill	`kimi-skill/SKILL.md`	Moonshot Kimi
MCP Server	✅ MCP Protocol	`mcp_server/`	Any MCP client
Python CLI	✅ Standalone	N/A	N/A

Platform Comparison

Feature	Claude Code	Gemini CLI	Codex CLI	Aider	Continue	Trae IDE	Kimi CLI
Cloud AI	Anthropic	Google	OpenAI	Any	Any	Multi	Moonshot
Local models	—	—	—	Ollama	Ollama	—	—
Skill system	SKILL.md	Extension	SKILL.md	—	Rules	Rules	SKILL.md
MCP support	Native	Native	Via Skills	Third-party	Native	—	—
Free tier	—	60 req/min	—	Self-hosted	Self-hosted	—	—

MCP Server Integrations

The vmware-aiops MCP server works with any MCP-compatible agent or tool. Ready-to-use configuration templates are in examples/mcp-configs/.

Agent / Tool	Local Model Support	Config Template	Integration Guide
Goose	✅ Ollama, LM Studio	`goose.json`	Guide
LocalCowork	✅ Fully offline	`localcowork.json`	Guide
mcp-agent	✅ Ollama, vLLM	`mcp-agent.yaml`	Guide
VS Code Copilot	—	`vscode-copilot.json`	Guide
Cursor	—	`cursor.json`	Guide
Continue	✅ Ollama	`continue.yaml`	Guide
Claude Code	—	`claude-code.json`	—

Fully local operation (no cloud API required):

# Aider + Ollama + vmware-aiops (via AGENTS.md)
aider --conventions codex-skill/AGENTS.md --model ollama/qwen2.5-coder:32b

# Any MCP agent + local model + vmware-aiops MCP server
# See examples/mcp-configs/ for your agent's config format

Installation

Step 0: Prerequisites

# Python 3.10+ required
python3 --version

# Node.js 18+ required for Gemini CLI and Codex CLI
node --version

Step 1: Clone & Install Python Backend

All platforms share the same Python backend.

git clone https://github.com/zw008/VMware-AIops.git
cd VMware-AIops
python3 -m venv .venv
source .venv/bin/activate
pip install -e .

Step 2: Configure

mkdir -p ~/.vmware-aiops
cp config.example.yaml ~/.vmware-aiops/config.yaml
# Edit config.yaml with your vCenter/ESXi targets

Set passwords via .env file (recommended):

# Use the template
cp .env.example ~/.vmware-aiops/.env

# Edit and fill in your passwords, then lock permissions
chmod 600 ~/.vmware-aiops/.env

Security note: Prefer .env file over command-line export to avoid passwords appearing in shell history. The .env file should have chmod 600 (owner-only read/write).

Password environment variable naming convention:

VMWARE_{TARGET_NAME_UPPER}_PASSWORD
# Replace hyphens with underscores, UPPERCASE
# Example: target "home-esxi" → VMWARE_HOME_ESXI_PASSWORD
# Example: target "prod-vcenter" → VMWARE_PROD_VCENTER_PASSWORD

Security Best Practices

NEVER hardcode passwords in scripts or config files
NEVER pass passwords as command-line arguments (visible in ps)
ALWAYS use ~/.vmware-aiops/.env with chmod 600
ALWAYS configure connections via config.yaml — credentials are loaded from .env automatically
Config File Contents: config.yaml stores target hostnames, ports, and a reference to the .env file. It does not contain passwords or tokens. All secrets are stored exclusively in .env
TLS: Enabled by default. Disable only for ESXi hosts with self-signed certificates in isolated lab environments
Webhook: Disabled by default. When enabled, sends monitoring summaries to your own configured URL only — payloads contain no credentials, IPs, or PII, only aggregated alert metadata. No data sent to third-party services
Least Privilege: Use a dedicated vCenter service account with minimal permissions. For monitoring-only use cases, prefer the read-only VMware-Monitor
Prompt Injection Protection: All vSphere-sourced content is truncated, stripped of control characters, and wrapped in boundary markers before output
Code Review: We recommend reviewing the source code and commit history before deploying in production
Production Safety: For production environments, use the read-only VMware-Monitor instead. AI agents can misinterpret context and execute unintended destructive operations — real-world incidents have shown that AI-driven infrastructure tools without proper isolation can delete production databases and entire environments. VMware-Monitor eliminates this risk at the code level: no destructive functions exist in its codebase

Step 3: Connect Your AI Tool

Choose one (or more) of the following:

Option A: Claude Code (Marketplace)

Method 1: Marketplace (recommended)

In Claude Code, run:

/plugin marketplace add zw008/VMware-AIops
/plugin install vmware-ops

Then use:

/vmware-ops:vmware-aiops
> Show me all VMs on esxi-lab.example.com

Method 2: Local install

# Clone and symlink
git clone https://github.com/zw008/VMware-AIops.git
ln -sf $(pwd)/VMware-AIops ~/.claude/plugins/marketplaces/vmware-aiops

# Register marketplace
python3 -c "
import json, pathlib
f = pathlib.Path.home() / '.claude/plugins/known_marketplaces.json'
d = json.loads(f.read_text()) if f.exists() else {}
d['vmware-aiops'] = {
    'source': {'source': 'github', 'repo': 'zw008/VMware-AIops'},
    'installLocation': str(pathlib.Path.home() / '.claude/plugins/marketplaces/vmware-aiops')
}
f.write_text(json.dumps(d, indent=2))
"

# Enable plugin
python3 -c "
import json, pathlib
f = pathlib.Path.home() / '.claude/settings.json'
d = json.loads(f.read_text()) if f.exists() else {}
d.setdefault('enabledPlugins', {})['vmware-ops@vmware-aiops'] = True
f.write_text(json.dumps(d, indent=2))
"

Restart Claude Code, then:

/vmware-ops:vmware-aiops

Submit to Official Marketplace

This plugin can also be submitted to the Anthropic official plugin directory for public discovery.

Option B: Gemini CLI

# Install Gemini CLI
npm install -g @google/gemini-cli

# Install the extension from the cloned repo
gemini extensions install ./gemini-extension

# Or install directly from GitHub
# gemini extensions install https://github.com/zw008/VMware-AIops

Then start Gemini CLI:

gemini
> Show me all VMs on my ESXi host

Option C: OpenAI Codex CLI

# Install Codex CLI
npm i -g @openai/codex
# Or on macOS:
# brew install --cask codex

# Copy skill to Codex skills directory
mkdir -p ~/.codex/skills/vmware-aiops
cp codex-skill/SKILL.md ~/.codex/skills/vmware-aiops/SKILL.md

# Copy AGENTS.md to project root
cp codex-skill/AGENTS.md ./AGENTS.md

Then start Codex CLI:

codex --enable skills
> List all VMs on my ESXi

Option D: Aider (supports local models)

# Install Aider
pip install aider-chat

# Install Ollama for local models (optional)
# macOS:
brew install ollama
ollama pull qwen2.5-coder:32b

# Run with cloud API
aider --conventions codex-skill/AGENTS.md

# Or with local model via Ollama
aider --conventions codex-skill/AGENTS.md \
  --model ollama/qwen2.5-coder:32b

Option E: Continue CLI (supports local models)

# Install Continue CLI
npm i -g @continuedev/cli

# Copy rules file
mkdir -p .continue/rules
cp codex-skill/AGENTS.md .continue/rules/vmware-aiops.md

Configure ~/.continue/config.yaml for local model:

models:
  - name: local-coder
    provider: ollama
    model: qwen2.5-coder:32b

Then:

cn
> Check ESXi health and alarms

Option F: Trae IDE

Copy the rules file to your project's .trae/rules/ directory:

mkdir -p .trae/rules
cp trae-rules/project_rules.md .trae/rules/project_rules.md

Trae IDE's Builder Mode reads .trae/rules/ Markdown files at startup.

Note: You can also install Claude Code extension in Trae IDE and use .claude/skills/ format directly.

Option G: Kimi Code CLI

# Copy skill file to Kimi skills directory
mkdir -p ~/.kimi/skills/vmware-aiops
cp kimi-skill/SKILL.md ~/.kimi/skills/vmware-aiops/SKILL.md

Option H: MCP Server (Smithery / Glama / Claude Desktop)

The MCP server exposes VMware operations as tools via the Model Context Protocol. Works with any MCP-compatible client (Claude Desktop, Cursor, etc.).

# Run via uvx (recommended — works with uv tool install)
uvx --from vmware-aiops vmware-aiops-mcp

# With a custom config path
VMWARE_AIOPS_CONFIG=/path/to/config.yaml uvx --from vmware-aiops vmware-aiops-mcp

Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "vmware-aiops": {
      "command": "uvx",
      "args": ["--from", "vmware-aiops", "vmware-aiops-mcp"],
      "env": {
        "VMWARE_AIOPS_CONFIG": "/path/to/config.yaml"
      }
    }
  }
}

Install via Smithery:

npx -y @smithery/cli install @zw008/VMware-AIops --client claude

Option I: Standalone CLI (no AI)

# Already installed in Step 1
source .venv/bin/activate

vmware-aiops vm power-on my-vm --target home-esxi
vmware-aiops deploy ova ./ubuntu.ova --name my-vm --target home-esxi
vmware-aiops datastore browse datastore1 --target home-esxi

Update / Upgrade

Already installed? Re-run the install command for your channel to get the latest version:

Install Channel	Update Command
ClawHub	`clawhub install vmware-aiops`
Skills.sh	`npx skills add zw008/VMware-AIops`
Claude Code Plugin	`/plugin marketplace add zw008/VMware-AIops`
Git clone	`cd VMware-AIops && git pull origin main && uv pip install -e .`
uv	`uv tool install vmware-aiops --force`

Check your current version: vmware-aiops --version

Chinese Cloud Models

For users in China who prefer domestic cloud APIs or have limited access to overseas services.

DeepSeek

Cost-effective, strong coding capability.

# Set DeepSeek API key (get from https://platform.deepseek.com)
export DEEPSEEK_API_KEY="your-key"

# Run with Aider
aider --conventions codex-skill/AGENTS.md \
  --model deepseek/deepseek-coder

Persistent config ~/.aider.conf.yml:

model: deepseek/deepseek-coder
conventions: codex-skill/AGENTS.md

Qwen (Alibaba Cloud)

Alibaba Cloud's coding model, free tier available.

# Set DashScope API key (get from https://dashscope.console.aliyun.com)
export DASHSCOPE_API_KEY="your-key"

aider --conventions codex-skill/AGENTS.md \
  --model qwen/qwen-coder-plus

Or via OpenAI-compatible endpoint:

export OPENAI_API_BASE="https://dashscope.aliyuncs.com/compatible-mode/v1"
export OPENAI_API_KEY="your-dashscope-key"

aider --conventions codex-skill/AGENTS.md \
  --model qwen-coder-plus-latest

Doubao (ByteDance)

export OPENAI_API_BASE="https://ark.cn-beijing.volces.com/api/v3"
export OPENAI_API_KEY="your-ark-key"

aider --conventions codex-skill/AGENTS.md \
  --model your-doubao-endpoint-id

With Continue CLI

Configure ~/.continue/config.yaml:

# DeepSeek
models:
  - name: deepseek-coder
    provider: openai-compatible
    apiBase: https://api.deepseek.com/v1
    apiKey: your-deepseek-key
    model: deepseek-coder

# Qwen
models:
  - name: qwen-coder
    provider: openai-compatible
    apiBase: https://dashscope.aliyuncs.com/compatible-mode/v1
    apiKey: your-dashscope-key
    model: qwen-coder-plus-latest

Local Models (Aider + Ollama)

For fully offline operation — no cloud API, no internet, full privacy.

Aider + Ollama + local Qwen/DeepSeek is ideal for air-gapped environments.

Step 1: Install Ollama

# macOS
brew install ollama

# Linux — download from https://ollama.com/download and install manually
# See https://github.com/ollama/ollama for platform-specific instructions

Step 2: Pull a model

Model	Command	Size	Note
Qwen 2.5 Coder 32B	`ollama pull qwen2.5-coder:32b`	~20GB	Best local coding model
Qwen 2.5 Coder 7B	`ollama pull qwen2.5-coder:7b`	~4.5GB	Low-memory option
DeepSeek Coder V2	`ollama pull deepseek-coder-v2`	~8.9GB	Strong reasoning
CodeLlama 34B	`ollama pull codellama:34b`	~19GB	Meta coding model

Hardware: 32B → ~20GB VRAM (or 32GB RAM for CPU). 7B → 8GB RAM.

Step 3: Run with Aider

pip install aider-chat
ollama serve

# Aider + local Qwen (recommended)
aider --conventions codex-skill/AGENTS.md \
  --model ollama/qwen2.5-coder:32b

# Aider + local DeepSeek
aider --conventions codex-skill/AGENTS.md \
  --model ollama/deepseek-coder-v2

# Low-memory option
aider --conventions codex-skill/AGENTS.md \
  --model ollama/qwen2.5-coder:7b