Sign inTry it

Benchmark: search products on Walmart with an AI agent

Wield's recipe found products on Walmart in 9 of 9 runs, with a median of 2.1 s and no model cost. A Claude browser agent passed 0 of 9 runs, with a median of 84.5 s and $0.12 per run. The benchmark flagged all 9 agent runs as blocked by the site. These results are preliminary, as of 7 November 2026.

Updated

Live recipewalmart.com/search-productsv1

Walmart Search Walmart products with price, rating and availability

Read-only, no loginRuns in your browser only

Parameters (6)
include_sponsoredboolean
max_priceintegerUSD
min_priceinteger
pageinteger
querystringrequiredWhat to search for, e.g. 'air fryer'
sortstring "best_match" | "price_low" | "price_high" | "best_seller" | "rating_high"
Use it with your agent
Measured on the live site
MetricWield recipeBrowser Use + Claude Sonnet 5.5
Success100%0%
Median time2.1 s85 s
Cost per run$0$0.122

9 runs each, 7 November 2026. How we measure

The task

Each side ran the same three searches. A run passes when it returns at least five products with a name and a price.

QueryParameters
q1query = air fryer, sort = price_low, max_price = 80
q2query = coffee maker
q3query = lego

Results

Wield recipeBrowser agent
Runs passing the check9 of 90 of 9
Median time per run2.1 s (2,147 ms)84.5 s (84,518 ms)
Model cost per run$0$0.12
Requests per run (median)1559 (to the site: 7)
Blocked by the site0 runs9 of 9 runs

The browser agent took a median of 11 steps per run, with 37,558 input tokens and 1,907 output tokens. Wield had no failed runs. Its fastest run took 2.0 s and its slowest 7.3 s.

Why the results differ

The browser agent ran in a fresh headless Chromium with an empty profile, the way cloud agents run. The log holds no HTTP 401, 403 or 429 for these runs. Our recipe notes name PerimeterX (HUMAN) as the bot protection on walmart.com.

Wield sends one search request from a tab in the user's own browser, at human pace. The recipe does not run a model, so it uses no tokens.

Caveats

  • Small sample. There are at most 9 runs per side, and the benchmark is still running. It is paced at about 30 site requests per hour.
  • Blocked means the benchmark's flag. It covers an HTTP 401, 403 or 429 from the site, or the agent reporting a block. Here the agent reported navigation failures.
  • Human pace only. Wield does not solve bot protection checks. If Walmart shows a check, Wield stops with site_challenge and the user solves it in their own tab.

Terms of Walmart

Walmart's terms prohibit automated access and scraping. The recipe makes one search page at a time that the user starts, from the user's own browser. Buying stays on walmart.com. Check the site's terms before you use any automation.

FAQ

Does Walmart block browser agents?

In our runs, yes. A Claude browser agent in a fresh headless Chromium was flagged as blocked in all 9 runs. Wield's recipe passed 9 of 9. The result comes from one agent setup on one date.

How fast is the Wield recipe for search products on Walmart?

In our benchmark as of 7 November 2026, the median run took 2.1 s. The fastest run took 2.0 s and the slowest 7.3 s.

Can I rerun this benchmark?

The method and per-task rows are on the benchmark page. The raw logs hold destinations, timing, tokens and check results, not page contents.