Skip to main content
Glama

One Jev-decided step

browser_act

Delegates one browser action from a one-sentence instruction; Jev selects and executes the click, type, select, scroll, wait, or key press, then returns status and a fresh observation.

Instructions

Delegate exactly ONE browser action to Jev (click, type, select, scroll, back, wait, press Enter, or stop). Give a one-sentence instruction; Jev picks the operation and target element; code validates, executes, and re-observes. Paid: a fraction of a cent (~$0.0002, more on very dense pages). Returns status, the executed step with its confidence, usage.est_cost_usd, a fresh observation, and — when a text field needs content — status 'needs_text'/'needs_value' with the ref to fill via browser_type/browser_select, then call browser_act again. Pass field text via values; value contents are never sent to Jev.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "instruction": "<instruction>"
    },
    "name": "browser_act"
  }
}

Input Schema

NameRequiredDescriptionDefault
valuesNoField label → text to type. Keys should match labels from the observation, e.g. {"Search": "espresso"}. Contents never go to Jev.
instructionYesOne sentence: the next action, e.g. "Click the Sign in link." or "Scroll down to the pricing table."

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.4/5.0
Behavior5/5

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

With no annotations, the description carries the full burden and does so: it discloses the paid cost (~$0.0002, higher on dense pages), the validation/execute/re-observe cycle, the returned fields (status, executed step with confidence, usage.est_cost_usd, fresh observation), and the needs_text/needs_value states. It also states a privacy guarantee — value contents are never sent to Jev.

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-loaded with the single-action constraint and the instruction format, then cost, then return shape, then the field-text workflow — a logical order. The prose is dense and parenthetical-heavy, but nearly every clause carries information an agent needs.

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

Completeness5/5

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

No output schema and no annotations exist, so the description must cover returns and safety — and it names the response fields, the cost model, and the follow-up states. The nested `values` object is also explained in workflow terms, leaving nothing essential missing for correct invocation.

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 description coverage is 100%, and the schema already documents both instruction and the values map including the label-matching rule and the 'never go to Jev' note. The description largely restates those points; its added value is tying `values` to the needs_text/needs_value follow-up flow rather than explaining parameter syntax.

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?

States a specific verb and resource — delegate exactly ONE browser action — and enumerates the covered operations (click, type, select, scroll, back, wait, press Enter, stop). It also distinguishes the delegation path from the deterministic siblings browser_type/browser_select when a text field needs content.

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 clear operating context: exactly one action per call, a one-sentence instruction, and an explicit follow-up loop (needs_text/needs_value → browser_type/browser_select → call browser_act again). It stops short of saying when to prefer a deterministic sibling like browser_click or browser_press outright instead of delegating to Jev.

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