Zilliz MemSearch Milvus Hybrid Vector Search Driver

@zilliztech/memsearch · v1.2.0

An enterprise Milvus and Zilliz Cloud vector driver for DeepSeek Harness. Delivers billion-scale hybrid vector search, scalar metadata filtering, partition management, and streaming indexing.

MilvusZilliz-CloudVector-DatabaseHybrid-Search
GitHub Stars
36 K+
+12.4% this month
Monthly Downloads
138 K+
Monthly registry pulls
Reach Score
98.5/ 100
Top Tier Ecosystem
Runtime
Cordis v3+
Node 18+ / Bun / Deno

Installation & Integration

CLI one-click launch, package managers, and Cordis integration

bash
manager:
$ pnpm add @zilliztech/memsearch

Architecture

Zilliz MemSearch scales DeepSeek Harness agent memory to billions of vectors. As knowledge repositories expand beyond millions of documents, embedded SQLite vector extensions choke under memory pressure and sluggish search latencies. This driver bridges Cordis directly with distributed Milvus and Zilliz Cloud clusters. Supporting unified hybrid search—fusing scalar attribute filters (tenant ID, ACL permissions, date ranges) with dense vector cosine similarity—it retrieves the most salient memories from billions of chunks in under 10 milliseconds.

Architectural Principles & Constraints

01
Strict Type Isolation

Guaranteed by TypeScript compile-time contracts, inter-plugin event bus calls enjoy zero-drift safety.

02
Sub-Millisecond Hot Reload

Supports dynamic runtime mounting and graceful unloading without restarting the primary host process.

03
Deterministic State Machine

Embeds multi-phase execution lifecycle guards, preventing context loss during long-horizon reasoning.

04
Zero Native Build Dependencies

Designed for lightweight cross-platform environments, booting instantly across Node.js, Bun, and Deno.

Core Features

01
Billion-Scale Vector Capacity: Scales horizontally across distributed Milvus cloud clusters
02
Unified Hybrid Search: Simultaneous execution of complex Boolean scalar ACLs and vector similarity
03
Hardware GPU Acceleration: Native support for GPU-accelerated vector indexes for sub-millisecond queries
04
Dynamic Partition Sharding: Enforces physical data isolation across multi-tenant enterprise partitions

Core Workflow

01

Chunking & Embedding Stream

Transforms text chunks into dense embeddings enriched with scalar metadata.

02

Partitioned Batch Ingest

Streams batched vectors into isolated Milvus partition collections via gRPC.

03

Hybrid Filtered Search

Executes hybrid searches combining Boolean scalar filters with cosine distance.

04

Reranking & Tenant Delivery

Reranks top candidates, delivering results strictly within tenant security walls.

Configuration Parameters Reference (YAML / JSON)

ParameterTypeDefaultDescription
milvusEndpointstringhttps://in01-xyz.serverless.gcp-us-west1.cloud.zilliz.comMilvus or Zilliz Cloud cluster endpoint
collectionNamestringdsh_agent_memoriesTarget Milvus collection identifier
metricTypestringCOSINEDistance calculation metric

Use Cases

Enterprise Production Agent

Relies on microkernel lifecycle guards and fault-tolerant state machines for continuous reliability.

SWE-bench Benchmark Evaluation

Native integration with SWE-bench workflows, automatically capturing diffs and verification metrics.

Autonomous Code Refactoring

Separates reasoning from tool actions to independently locate and refactor multi-file codebases.

Cross-Tool Workflow Automation

Safely orchestrates events across sandboxes to seamlessly link enterprise developer tooling.

Best Practices

01
Sandbox Permission Guard

Strictly isolate sub-process calls and network scope; deploy within Docker containers in production.

02
Exponential Backoff Retries

Configure adaptive exponential retries with strict timeouts to mitigate upstream model rate limits.

03
Session State Checkpointing

Persist state machine snapshots to survive hardware interruptions and resume instantly without loss.

04
Full Trajectory Audit Logs

Enable full trace logging, aggregating reasoning thought streams and tool I/O into your observability hub.

FAQ

Q1:How to handle timeouts in long-running autonomous tasks?

Increase the timeout parameter inside your YAML configuration and dispatch periodic heartbeat signals across the Cordis event bus. For long-running tool execution and model reasoning, configure persistent session snapshotting so suspended tasks can safely resume their exact context after interruptions, preventing the kernel from recycling active agent sessions prematurely.

Q2:How to capture and stream the model reasoning thought process?

The harness runtime natively provides end-to-end streaming hooks while its state machine automatically intercepts and strips <think> reasoning tags from model responses. Subscribe directly to the onThink event listener to consume live reasoning token streams in real-time, delivering typewriter animation to the user interface while persisting full trajectories for audit compliance.

Q3:How to resolve dependency conflicts across multiple plugins?

Cordis microkernel uses directed acyclic graph topological sorting to dynamically resolve plugin dependencies. When shared services or runtime versions conflict, assign distinct isolated namespaces at the application entrypoint. Leveraging context injection alongside lazy activation ensures dependencies load strictly on-demand when tools are triggered, maintaining system stability and preventing memory bloat.

Q4:How to enforce permissions and sandbox isolation in production?

Combine isolated container sandboxing with fine-grained capability checks to prevent plugins from accessing sensitive files or unauthorized external networks. Enforce explicit runtime system call whitelisting through the microkernel, executing all third-party tool scripts inside isolated ephemeral containers to block arbitrary code execution and eliminate security privilege escalation risks entirely.

Recommended Ecosystem Plugins

Explore related Cordis extensions designed to work synergistically