Skip to main content
Glama

site_verify

Runs one browser flow across desktop and phone viewports and returns one structured verdict: every named step, page/console/network health, expected request outcomes, both screenshot paths or URLs, and a non-blocking accessibility/layout line per viewport covering WCAG AA text contrast, horizontal overflow, and tap targets below the 24×24px WCAG 2.2 AA minimum (24–44px is listed as a note). Authenticated or seeded flows require project_id and a project-owned origin. An absent action list produces a load/health/two-screenshot check. Each viewport uses a fresh browser that closes inside the call, including on failure.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "site_verify"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlNoPublic live URL. Without project_id, external URLs support inspection and screenshots only. With project_id, navigation and action targets must remain on an authorized origin owned by that project. Normal external assets and APIs loaded by the owned app remain available under backend policy.
authNoOptional logged-in app user for every viewport: { user_id }. Requires an owned project target and mints an audited one-hour session.
actionsNoConcise browser actions requiring project_id and a project-owned navigation origin: {click}, {fill,value}, {upload,file|fs_path,name?,type?} (fs_path: a project file up to 50 MB per call), {select,value}, {wait}, {expect:{selector,text?|visible?|count?}}, {screenshot}, {eval}, or {goto:"/path"}. Execution stops at the first failed action and names its 1-based step in the verdict.
cookiesNoOptional existing cookies seeded before navigation for every viewport. Requires url + project_id and is origin-scoped.
headersNoOptional request headers seeded for every viewport. Requires url + project_id and is origin-scoped.
budget_msNoExecution budget in milliseconds, including startup (default 180000). A transport timeout has an unknown outcome; actions are not replayed automatically.
viewportsNoViewport list. Defaults to ["desktop", "mobile"] (1280×800 and 390×844). Each item is a preset or { label, width, height }.
project_idNoOwned project authority. Required for actions, eval, authentication, and session seeds; it also enables screenshots in project files.
visible_onlyNoWhen true, action selectors match visible elements only. Defaults to false, matching the existing browser actions contract.
local_storageNoOptional existing session values seeded before navigation for every viewport. Requires url + project_id and is origin-scoped.
expect_requestsNoExpected request outcomes as { path, status }. A matching refusal is excluded from console/network failures; a missing or different status fails verification.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A3.6/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=false, openWorldHint=true, idempotentHint=false and destructiveHint=false, so the safety profile is covered. The description adds genuinely useful behavior beyond that: a fresh browser per viewport that closes inside the call even on failure, execution halting at the first failed action with its 1-based step named in the verdict, and timeouts having an unknown outcome with no automatic replay. The gap keeping this at 3 is that it never resolves the tension between readOnlyHint=false and the fact that the tool takes no destructive action, nor does it say what state, if any, it mutates.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the outcome (one flow, both viewports, one structured verdict) and then enumerates the verdict contents, prerequisites, and lifecycle guarantees. It is dense and slightly run-on, but nearly every clause carries information an agent needs; the qualifying parenthetical about 24–44px tap targets is the one element that could be trimmed.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description carries the burden of explaining returns, and it does so concretely: per-step results, page/console/network health, expected request outcomes, both screenshot paths or URLs, and a per-viewport accessibility/layout line. Combined with the disclosed browser lifecycle and budget semantics, an agent has enough to call this correctly, though it could state when to reach for it over the sibling check tools.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline of 3 applies: the schema already documents all eleven parameters, including the action shapes and viewport presets. The description adds the relationship between project_id and authorized origins and the default behavior when actions is omitted, which is helpful but marginal over what the schema already says.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: runs one browser flow across desktop and phone viewports and returns a structured verdict. It clearly describes what the call produces, but never names or contrasts with sibling tools like site_check, browser, or site_check_status, so an agent deciding between them gets no routing help from the definition itself.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives real usage context: authenticated or seeded flows require project_id and a project-owned origin, and an absent action list degrades to a load/health/two-screenshot check. It never explicitly says when to prefer this over site_check or site_check_status, but the preconditions and the empty-actions behavior are clear enough to guide invocation.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources