Skip to content
Trending storyUpdating

ToolForte: What I Learned Integrating 360 Browser Tools into MCP

2 reports1 reporting sourceUpdated 13 hours ago

Understand the story

AI synthesis

The author of ToolForte wired 360 in-browser tools into an MCP server, starting with 111 tools exposed and later growing to 180. In a post-mortem on DEV Community, the author said the oversized tool list was eating into the context window and the agent kept picking the wrong tool, so they switched to showing only 14 tools by default and making the other 166 discoverable through search. In follow-up coverage, the author looked back at connecting 11 MCP servers to a coding agent in February: an empty session already spent 34,000 tokens on the tool list alone, with Datadog accounting for over two hundred tools by itself, which slowed the agent down and caused frequent wrong picks.

Generated by AI from reports · Updated 13 hours ago

Developments

2 developments
  1. Oct 6, 15:17 · 1 report
    Using Skills to Fix Context Waste from a Bloated MCP Tool List
    DEV Community · MCP:Skills Are Not Tools: Why I Gave My Coding Agent 11 MCP Servers and It Fell Apart
  2. Oct 5, 22:34 · 1 report
    ToolForte MCP server: launch and post-mortem
    DEV Community · MCP:ToolForte 为 360 个浏览器工具搭建 MCP server:哪些地方出了问题

Report timeline

Follow the reports to explore different perspectives.

10/6
  1. DEV Community · MCPSelected
    Skills Are Not Tools: Why I Gave My Coding Agent 11 MCP Servers and It Fell Apart

    In February I hooked up 11 MCP Servers to my coding agent. Just the tool list in an empty session ate up 34,000 tokens, with Datadog alone contributing over two hundred tools. The agent got slower and kept picking the wrong tools.

10/5
  1. DEV Community · MCP
    ToolForte 为 360 个浏览器工具搭建 MCP server:哪些地方出了问题

    ToolForte 作者把站内 360 个浏览器小工具的 111 个(后增至 180 个)暴露为 MCP 工具,结果发现列表过大挤占上下文窗口、智能体频繁选错工具,改为默认只展示 14 个、其余 166 个靠搜索查找。

Interest in this story

Current interest 7·Peak within comparable coverage 10(Oct 6, 17:00)·Change over 24 hours within comparable coverage –

02.557.51010/616:0010/620:0010/700:0010/704:00

The trend compares the same participants under continuous, complete observation, so its coverage may be narrower than the current score. Hover or tap to view hourly interest; use the left and right arrow keys to navigate.