How to Add Craigslist Search to Codex

Codex CLI is built for fast terminal workflows, but it can’t browse classified listings. When you need to find used equipment, check local prices, or search for workspace rentals, you have to do it yourself. Adding Craigslist Search through AgentPatch gives Codex direct access to listings.

Why This Matters

Whether you’re sourcing hardware for a project, looking for used networking gear, or pricing out equipment before buying new, Craigslist is often the first place to check. But switching between your terminal and a browser breaks your flow.

With Craigslist Search available, Codex can search listings in any city and category, returning structured results you can work with — titles, prices, locations, and direct links. No browser tabs, no copy-pasting.

Setup

Add AgentPatch to ~/.codex/config.toml:

[mcp_servers.agentpatch]
url = "https://agentpatch.ai/mcp"
bearer_token_env_var = "AGENTPATCH_API_KEY"

Then set your API key:

export AGENTPATCH_API_KEY=your_api_key

Replace your_api_key with your actual key from the AgentPatch dashboard. Codex discovers all AgentPatch tools automatically on next start.

Example

You’re setting up a home lab and need a used server rack:

“Search Craigslist in the New York area for a server rack”

Codex searches Craigslist instead of guessing:

Searching Craigslist newyork for "server rack"...

Found 12 listings:

1. 42U Server Rack Cabinet w/ rails - $350 (brooklyn)
2. Dell PowerEdge R720 + 24U Rack - $500 (jersey city)
3. Open Frame 4-Post Server Rack 25U - $175 (queens)

Want me to narrow these down by price or look in a different city?

Codex searched, parsed the results, and gave you structured data. You can follow up to filter or expand the search without leaving your session.

Wrapping Up

Connecting AgentPatch to Codex gives it access to Craigslist Search during any session. Beyond classified listings, the same connection exposes every other tool on the marketplace — Google Search, Amazon Search, email, and more. See what’s available at agentpatch.ai.

← Back to blog