Toolgz – cut LLM tool-definition tokens ~80% without hurting accuracy
toolgz Your agent spends 30–50k tokens of context on tool definitions before the user types a word. toolgz gets ~80% of it back. 420-run cross-provider sweep · 4 frontier models · zero runtime dependencies · generated before/after npm install toolgz The problem You connect a few MCP servers. Every tool is a JSON Schema with a sentence of prose per parameter.
- ▪toolgz Your agent spends 30–50k tokens of context on tool definitions before the user types a word. toolgz gets ~80% of it back.
- ▪420-run cross-provider sweep · 4 frontier models · zero runtime dependencies · generated before/after npm install toolgz The problem You connect a few MCP servers.
- ▪Every tool is a JSON Schema with a sentence of prose per parameter.
Hacker News (AI / LLM) files mainly under ai. We currently carry 2,292 of its stories.
Opening excerpt (first ~120 words) tap to expand
toolgz Your agent spends 30–50k tokens of context on tool definitions before the user types a word. toolgz gets ~80% of it back. 420-run cross-provider sweep · 4 frontier models · zero runtime dependencies · generated before/after npm install toolgz The problem You connect a few MCP servers. Each ships 20–50 tools. Every tool is a JSON Schema with a sentence of prose per parameter. That block renders at the front of every single request. A realistic tool definition is ~420 tokens, and roughly 400 of them are prose the model doesn't need in order to pick correctly. Fifty tools is 20k tokens. A hundred is 40k. Prompt caching makes those tokens cheap. It does not make them take up less room. Reclaiming the room is what this does.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.