Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine
TiDB is built for agentic workloads that grow unpredictably, with ACID guarantees and native support for transactions, analytics, and vector search. No data silos. No noisy neighbors. No infrastructure ceiling.
ACRYL - Agent Context Relay Yielding Lifecycles. One persistent workspace, one canonical context, any coding agent.
Cross-agent plugin for manually excluding and editing AI conversation context, with search, filtering, hide/restore, undo, and original Session preservation.
A minimal coding-agent harness built to measure what context management actually costs: constraint survival after compaction (8%β83%), default-FAIL verification against false completions, and KV-cache prefix design (21%β90%). Stdlib-only, real deepseek-chat runs