Just trying things
token-savior. The MCP server that turns Claude into the only coding agent hitting 100% on a real benchmark. -77% active tokens, -76% wall time, 0 losses across 96 tasks on Claude Opus 4.7. Structural code navigation + persistent memory. Works with every MCP client.
1.1ktsbench. The benchmark Token Savior runs at 100% on. 96 real coding tasks, paired runs, oracle-graded ground truth. Plain Claude scores 78.3%; Token Savior scores 192/192. Reproducible with one command.
6hermes-memory. Structured persistent memory for LLM agents — MCP-native, SQLite, zero infra
3oodrive-sign-agent. Complete Oodrive Sign API v4 integration skill for AI agents — auth, contracts, smart fields, gather, webhooks, TypeScript helpers, 17 pitfalls
1