Tag
9 pieces
We probed a single MCP server installed with one line of npx config. It exposes 372 tools and 335,954 bytes of JSON schema, roughly 42 percent of a 200K context window.
We denied reads of one file and calls to curl, then tried eight ways around it. Six were blocked, including cat, sed and a chained command. The two that were not still stopped and asked.
Herdr is not a terminal, it is a multiplexer that runs inside one. Across 181 runs it consumed 4.4MB of output in 0.059s, faster than the Ghostty window hosting it.
A three-token headless prompt sent 44,137 tokens of preamble and cost 31 cents. The identical command four seconds later cost 2.2 cents.
Four Claude models, a 10x price spread, and a default recommendation that is not the most expensive one. The selection guide also says tuning effort beats switching models.
A PreToolUse hook that blocks rm -rf took ten minutes to write and caught 2 of 6 ways to delete a directory. Worse, it failed open: broken, it exited 0 and let the delete through.
An audit of 77 Claude Code rule files found 22 loading on every session, and 10 of those were Chinese translations of the other 10. That is 4,265 tokens of duplicate instruction per session.
Claude Code budgets its skill listing at 1 percent of the context window. We counted 228 skills across four scopes totalling 50,606 characters, 25 times over budget on a 200K model.
A published inventory of ChatGPT Work lists 232 tool interfaces, 44 skills and 615,000 characters of skill source. Its browser skill deliberately withholds its own instructions, and the reason is arithmetic.