AGY Dedicated Inference Provider & Connection Pool

Official Core
@deepseek-ai/dsh-provider-agy · v2.5.0

Official AGY Cloud dedicated inference driver for DeepSeek Harness. Features HTTP/2 multiplexed persistent connections, binary streaming protocols, and ultra-low latency cluster routing.

Official-ProviderAGY-CloudConnection-PoolLow-Latency
GitHub Stars
134 K+
+12.4% this month
Monthly Downloads
581 K+
Monthly registry pulls
Reach Score
99.4/ 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 @deepseek-ai/dsh-provider-agy

Architecture

AGY Provider represents the enterprise-grade inference backbone for DeepSeek Harness. High-throughput agent pipelines repeatedly establishing short-lived HTTPS connections incur heavy TLS handshake overhead and TCP socket exhaustion. This driver re-engineers the network layer with persistent HTTP/2 connection pooling and zero-copy binary streaming. Designed to interface directly with dedicated AGY enterprise clusters, it minimizes Round-Trip Time (RTT) and effortlessly sustains thousands of concurrent token streams.

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
Multiplexed HTTP/2 Connection Pool: Channels hundreds of concurrent streams over persistent sockets
02
Ultra-Low Jitter Dedicated Routing: Reduces network latency by up to 50% vs public endpoints
03
Adaptive Client Backpressure: Throttles high-speed token generation to protect slower consumer sinks
04
Private Dedicated Cluster Native: Switches seamlessly between public AGY Cloud and on-prem hardware

Core Workflow

01

Persistent Pool Warmup

Pre-warms multiplexed HTTP/2 sockets with target AGY clusters upon startup.

02

Zero-Copy Serialization

Serializes prompts and tool schemas into binary frames without memory copies.

03

Dedicated Pipe Streaming

Streams prompts and receives token chunks over low-jitter enterprise pipes.

04

Adaptive Backpressure

Dynamically throttles token emission to prevent client-side buffer overflows.

Configuration Parameters Reference (YAML / JSON)

ParameterTypeDefaultDescription
poolSizenumber16Persistent HTTP/2 socket pool capacity
connectTimeoutMsnumber5000Socket connection timeout in milliseconds
enableCompressionbooleantrueEnable Brotli/Gzip wire payload compression

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