jevpilot

browser_run

Drive a real Chrome browser toward one goal per call, returning either a verified result or a concrete question instead of guessing. Reuse the session to continue, observe, act, or close.

Instructions

Open a new browser session and work toward a goal. Jev acts fast on its own and hands back early with a question when unsure; continue with browser_act / browser_resume from the returned snapshot. Returns a session ID and a status; handoff statuses include a concrete question for the agent or user. Reuse the session with browser_resume, browser_observe, browser_act, and browser_close.

Input Schema

TableJSON Schema

Name	Required	Description
`url`	No	Initial HTTP(S) URL. Omit to start on a blank tab.
`goal`	Yes	Task to complete in the browser.
`budget`	No	Per-invocation limits.
`values`	No	Every piece of text the page needs typed (search terms, form fields) must be given here as { short field description: text }. Text inside the goal is never typed. Secrets use {secret_ref: 'env:NAME' or 'file:PATH', origins: [...] }.
`profile`	No	Configured engine profile name; defaults to the server engine.
`success`	No	Optional checks that are false now and become true only when the goal is done, e.g. url_matches for the result page. Checks that already hold on the start page are ignored.
`thresholds`	No	Per-session decision confidence thresholds (0 to 1).
`constraints`	No	Navigation allowlist and irreversible-action gate.
`navigation_timeout_ms`	No	Navigation timeout in milliseconds; defaults to 30000.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`url`	Yes
`title`	Yes
`trace`	Yes
`usage`	Yes
`reason`	Yes
`status`	Yes
`timing`	Yes
`details`	No
`session`	Yes
`question`	Yes
`snapshot`	Yes
`screenshot_path`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 29, 2026

TDQS

A3.7/5.0

Behavior4/5

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

With no annotations, the description carries the burden and does disclose meaningful behavior: Jev acts autonomously and 'hands back early with a question when unsure', plus the handoff-status semantics. This is genuine runtime behavior not expressed in the schema, though it omits any note on irreversible-action or auth implications (those live only in the schema).

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?

Purpose is front-loaded in the first sentence and the follow-up guidance is compact. It is slightly repetitive about the resume/act siblings, but every sentence carries useful routing information.

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?

For a complex autonomous tool with an output schema, the description covers the goal, the handoff model, and the continuation path. Return-format details are correctly left to the output schema, so the remaining gaps are minor.

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%, so the schema already documents all 9 parameters and their nested fields thoroughly. The description adds nothing parameter-specific, so the baseline 3 is appropriate.

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 ('Open a new browser session and work toward a goal'), which is clear and distinct from the resume/reuse flow. It names the session-reuse siblings but does not explicitly contrast with browser_navigate, which also concerns browsing.

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

Usage Guidelines3/5

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

It explicitly tells the agent how to continue after a handoff ('continue with browser_act / browser_resume from the returned snapshot') and that the session can be reused with several siblings. However, it never states when to choose browser_run versus browser_navigate, and gives no exclusions or prerequisites.

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

Deploy Server