MCP server for browser automation using Bright Data Scraping Browser via Playwright, enabling web navigation, interaction, and content capture
Server has 8 tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (browser_navigate, browser_click), which is good. However, descriptions are minimal (10-30 chars), parameter descriptions are sparse or missing, and output schemas are not documented. No error handling guidance, no parameter constraints (enums, ranges), and no indication of idempotency or side effects. The server lacks the depth expected for production use.
Click on an element
Get the HTML content of the current page
Go back to the previous page
Go forward to the next page
Navigate to a URL
Take a screenshot of the current page
Type text into an element
Descriptions are too short (10-40 chars) and lack context. Baseline is 194 chars. LLMs cannot determine when to select tools without fuller descriptions explaining WHAT, WHEN, and WHY.
Output schemas are not documented. Tools return ImageContent or TextContent but LLMs have no visibility into what fields to expect, breaking downstream tool chaining and forcing guesswork.
No error handling guidance. Tools catch errors and return isError=true with raw error text, but do not tell LLMs what to do next (retry, ask user, abort). Baseline pattern requires recovery hints.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 42 | 2026-07-28+ | v2 |
Wait for an element to be visible on the page
Parameter descriptions are missing or minimal. 'selector' appears in 4 tools but is never explained (CSS selector syntax, required format, examples). Baseline: 100% of A+ tools have param descriptions.
No indication of idempotency or side effects. browser_click and browser_type are destructive (WRITE risk) but descriptions do not warn LLMs. Agents need to know which calls are safe to retry.