Slimtoken
未认领A standalone, all-on-by-default token-optimization layer for LLM requests: minifies tools/system/messages (code-fence aware), dedups tool results, distills old turns, prunes to a budget, type-compresses output. Async proxy for Anthropic, OpenAI & Ollama + MCP server (stdio) + CLI. MIT; ships orjson/xxhash/tiktoken.
anthropicasynccontextllama-serverllmmcpollamaopenaiproxytoken-optimization