# Use Affogata Benchmarks with your agent.

[Affogata Benchmarks](https://benchmarks.affogata.net/) (`benchmarks.affogata.net`) registers 16 WebMCP tools. 11 of 16 tools declares `readOnlyHint`, so an agent can invoke them through `tools.call`. The registry runs the tool in a fresh headless page on benchmarks.affogata.net and returns untrusted site content.

This page mirrors the [WebMCP Registry](https://wmcp.ai/sites/benchmarks.affogata.net) listing for `benchmarks.affogata.net`, last live-checked by the registry at 2026-09-23T02:01:10.371Z. Run `ghostget webmcp sites.get --input '{"domain":"benchmarks.affogata.net"}' --json` for the current schema; the listing can drift between checks.

## Use Affogata Benchmarks with my agent

Ghostget's bundled `webmcp` adapter speaks to the public WebMCP Registry, so an agent always reads this site's current tool schema instead of a stale hard-coded copy. Install v0.18.35, sync bundled adapters once, then call the registry operations:

```
bun add --global https://github.com/hraness/ghostget/releases/download/v0.18.35/hraness-ghostget-0.18.35.tgz
ghostget adapter sync-bundled --json
```

**Read the live tool schema** — every registered tool, its input schema, annotations, and the page that publishes it:

```
ghostget webmcp sites.get --input '{"domain":"benchmarks.affogata.net"}' --json
```

**Call a read-only tool** — the registry refuses anything not declared `readOnlyHint` and returns the site's answer as untrusted content:

```
ghostget webmcp tools.call --input '{"domain":"benchmarks.affogata.net","tool":"benchmarks_compare_titles","input":"{}"}' --json
```

## Registered WebMCP tools

Affogata Benchmarks publishes 16 tools on benchmarks.affogata.net; 11 declare `readOnlyHint`.

| Tool | What it does | Callable through Ghostget |
| --- | --- | --- |
| `benchmarks_compare_in_ui` | Put titles side by side on the comparison page so the user can see the difference rather than read it. The default mode replaces the whole selection, which is right for 'compare X and Y' and wrong for 'add a competitor' - use mode 'add' to | Listed for discovery; the registry allows read-only calls only |
| `benchmarks_compare_titles` | Compare 2 to 5 titles side by side across score, release delta, weekly movement, category position, store rating and volatility, plus the topic clusters they share and the ones unique to each. Works across categories. | Callable read-only tool |
| `benchmarks_explain_methodology` | Explain what the score means, where the data comes from, what the score bands are, and what its limits are. Call this before presenting any number to a user so the framing is accurate. | Callable read-only tool |
| `benchmarks_export_dataset` | Return the raw benchmark rows as structured JSON so you can run your own analysis - correlations, cohort comparisons, charts, or a written report. Filter by category or industry to keep the payload small. | Callable read-only tool |
| `benchmarks_find_by_topic` | Search the topic clusters across every benchmarked title to find who is being praised or punished for a given theme - 'paywall', 'support', 'ads', 'bugs', 'pay to win'. This is the cross-category view a per-brand lookup cannot give you. | Callable read-only tool |
| `benchmarks_get_category_analytics` | Deep read on one category: average and median score, spread, leader and laggard, the strong/mixed/critical split, this week's gainers and droppers, and the topic clusters most often lifting or dragging titles in that category. | Callable read-only tool |
| `benchmarks_get_current_view` | Read what the page is showing right now: the route, the active filters, and the exact titles pinned in the comparison. The person at the keyboard drives the same page you do - they can remove a comparison chip, add a title or change a filte | Callable read-only tool |
| `benchmarks_get_movers` | Rank titles by week-over-week movement to find what changed since the last report - the fastest way to spot a release that just went wrong or a fix that landed. | Callable read-only tool |
| `benchmarks_get_overview` | Headline state of the whole benchmark: how many titles are strong, mixed or critical this week, the happiest and unhappiest customer bases, the biggest weekly mover, and the topics rising and falling across the whole corpus. | Callable read-only tool |
| `benchmarks_get_title` | Full customer-voice profile for one game or app: current score, release history, the topic clusters pushing it up and down, the Affogata insight, movement against its category, and store-review volume where a live pull exists. | Callable read-only tool |
| `benchmarks_list_categories` | List every benchmark category with its industry, title count, average customer-voice score and current leader. Start here when you do not yet know which category a brand belongs to. | Callable read-only tool |
| `benchmarks_list_titles` | List benchmarked games and apps with their current score, the change on their latest scored point (a release or a week, per the title cadence) and their week-over-week move. Filter by category, industry, free text or minimum score, sort by | Callable read-only tool |
| `benchmarks_open_category` | Switch the page to a benchmark category and show its titles. Use this whenever the user asks to see, open or switch to a category - the browser view follows along instead of the answer living only in chat. | Listed for discovery; the registry allows read-only calls only |
| `benchmarks_open_title` | Open the detail page for one game or app, showing its score history, topic clusters and movement. Use it to put a specific brand in front of the user after you have talked about it. | Listed for discovery; the registry allows read-only calls only |
| `benchmarks_reset_view` | Clear every filter, comparison and highlight and return the page to the full benchmark overview. Use it when the user says 'start over' or 'show everything again'. | Listed for discovery; the registry allows read-only calls only |
| `benchmarks_set_view` | Change what the browsing page shows: free-text search, a topic filter, an industry, a minimum score, the sort order, and grid or table layout. This is the agent equivalent of the on-page controls, so the user ends up looking at exactly the | Listed for discovery; the registry allows read-only calls only |

## What to expect

- **Read-only only.** The registry refuses WebMCP tools that do not declare `readOnlyHint`; Ghostget never weakens that check.
- **Untrusted results.** Tool output is site content, not instructions — treat it as data.
- **Live schemas.** `sites.get` always returns the schema the registry saw most recently, so agents adapt when Affogata Benchmarks changes its tools.
- **No account needed.** These calls are credential-free registry reads. For tools that need a session on benchmarks.affogata.net, browse the site itself or check back when the registry lists more tools.