MCP
-
An eBay MCP server spends 40% of my context window on its tool list.
I measured the MCP context cost of 22 server setups. One eBay server takes 40.6% of a 200k window if your client loads every tool. Here is what to keep.
-
This PDF MCP server’s redact button only paints a box. My card number stayed.
I ran PaperPilot, a local PDF MCP server with 76 tools, on fake invoices. Its redact tool leaves the card number in the file. Here is the test.
-
Kanzo is a one-developer kanban board for coding agents. One call cost 92k tokens.
I ran the Kanzo MCP server, a 4.9 MB kanban board for agents, through 92 task calls in five bursts: zero doubles. Then three things broke my trust.
-
MCP Skills: I Tried Them, and Verified Is Not Trusted
MCP skills ship in the official TypeScript SDK 2.4. I ran them, and a hostile server's prompt injection still came back verified: true. Here is why.
-
SolidCraft MCP: The Tool Said OK, the Part Had No Holes
SolidCraft MCP is an open-source CAD server for AI agents. I ran it: great design, one trap where ok:true meant nothing changed. How to guard against it.
-
Why I’m Betting on a Minimal Agent Harness
Pi 1.0's own release notes claim a 40% prompt-token cut from Codemode. Here's the arithmetic I'd run before trusting a minimal agent harness.
-
The Hidden Risk of Letting AI Manage Ad Campaigns
MCP's own specs disagree on whether it is stateful. Neither defines a record of what changed in your ad account — and an unattended run needs that record.