nocturnusai
MCPMCP server providing agent reasoning, memory, and token-optimized context management for AI applications.
npx -y nocturnusai-mcpThe context engineering engine for AI agents: send only what changed.
Measured on live APIs. 15-turn product support conversation. Real `usage.input_tokens` counts. Run it yourself.
Claude Opus 4: 5.7× reduction. Gemini 2.0 Flash: 10.0×. Full calculations.
Reproducible token reduction — benchmark in the repo, methodology published, run it against your own workload Deterministic inference — same query, same result, every time. No embedding drift, no cosine similarity lottery Truth maintenance — retract a fact, all derived conclusions auto-retract. No stale context, no hallucination on operational state Plugs into existing stacks — LangChain, LlamaIndex, CrewAI, AutoGen, MCP, Vercel AI SDK, OpenAI Agents SDK, Mastra Benchmarkable against naive re…
- Source
- community
- Known advisories
- 0
- Maintenance
- active
- License
- none
- Age
- 0 months