Skip to main content
Glama

eyes_see_browser

Inspect a browser page's title, headings, visible text, links, buttons, inputs, and forms, returning stable refs for interaction; optionally use a private read-only context or attach a screenshot.

Instructions

See the browser page: title, headings, visible text, links, buttons, inputs and forms. Interactive elements get stable refs (btn_1, input_2, link_3) usable by hands_*. If private=True: the URL is opened in a throwaway isolated context (fresh cookies/storage), read-only, refs omitted, context always closed — weaker local persistence, NOT anonymity. screenshot=True attaches ONE downscaled image of the page (vision mode, on request). Restricted pages (CAPTCHA, rate limit, paywall, access denial) are reported, never bypassed.

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": "eyes_see_browser"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlNo
privateNo
screenshotNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.4.0

TDQS

A4.6/5.0
Behavior5/5

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

With zero annotations the description carries the full load and does so well: it discloses that private mode is a fresh-cookie isolated context, read-only, refs omitted, always closed, and explicitly warns this is weaker local persistence, 'NOT anonymity'. It also states restricted pages are reported and never bypassed, and that screenshot adds exactly one downscaled image.

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

Conciseness5/5

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

Dense but well-organized; every clause adds operative detail an agent needs.

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?

No output schema and no annotations, yet the description covers return content, ref engineering, both boolean flags, isolation semantics and failure handling. The residual gap is url behavior when omitted and whether a browser must already be attached.

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

Parameters4/5

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

Schema coverage is 0%, so the description must supply meaning, and it does for private and screenshot in detail, including the ref omission and context-closing consequences. The url parameter is only implied ('the URL is opened'), and its null default (presumably current page) is never explained, leaving one gap.

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

Purpose5/5

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

The description names a specific verb and resource — 'See the browser page' — and enumerates exactly what is returned (title, headings, visible text, links, buttons, inputs, forms). The ref scheme (btn_1, input_2, link_3) further pins down the output shape so an agent knows this is a structured page-reading tool, not a screenshot or desktop tool.

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?

It routes the agent clearly: refs are for hands_* follow-up actions, private=True selects a throwaway isolation mode, and screenshot=True is described as an on-request vision mode. What's missing is an explicit statement of when to prefer this over its sibling eyes_see_desktop or when attach_browser is a prerequisite.

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