Scroll a page by a specified number of pixels up or down in a background tab, letting lazy-loaded and infinite lists continue loading.
Scrolls the web page by y pixels down (down) or up (up). A background tab renders while scrolled: lazy and infinite lists (X, Reddit, YouTube...) keep loading.
| Name | Required | Description | Default |
|---|---|---|---|
| y | No | ||
| tab_id | Yes | Your tab's id (from browser_new_tab, or a tab adopted with browser_adopt_tab). | |
| direction | No | down |
Changes observed during successful MCP inspections.
v2.31.6Output schema / (root)Previous value: -{
- "properties": {
- "result": {
- "title": "Result",
- "type": "string"
- }
- },
- "required": [
- "result"
- ],
- "title": "browser_scrollOutput",
- "type": "object"
-}New value: +nullv0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare openWorldHint=true and destructiveHint=false, so safety is partly covered. The description adds a real behavioral trait beyond annotations — that scrolling causes lazy/infinite lists to keep loading — but the line 'A background tab renders while scrolled' is opaque and never clarifies whether scrolling affects page state, clamps at page end, or triggers network activity the caller should wait on.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, zero padding, with the core action and direction semantics front-loaded ahead of the secondary lazy-loading note.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 3-parameter tool with annotations and no output schema, this covers the essential call semantics and the notable side effect. It could still say what the call returns (new scroll offset, confirmation) and whether tab_id defaults to a particular tab.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is only 33% (only tab_id is documented), but the description compensates by defining y as pixel offset and direction as down/up. This matters because direction has no enum in the schema, so the description is the only place the legal values are stated; it still omits the default y=400 and the 'down' default.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (scrolls), the resource (the web page), the unit (y pixels) and the direction axis, so the action is unambiguous. No sibling tool in the list does scrolling, and the name plus first sentence together remove any doubt about what it affects.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The lazy/infinite-list sentence implies the use case (triggering more content to load) but stops short of saying when to prefer this over alternatives like browser_press_key (PageDown/End), browser_find, or browser_wait_for, and gives no exclusions or preconditions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.