DSH Dynamic Router Intelligent Routing & Cost Balancer

@yjh051108/dsh-routing-suite · v1.2.0

An adaptive task-complexity model router for DeepSeek Harness. Automatically routes lightweight tasks to fast small models while escalating deep logical reasoning to DeepSeek-R1, saving costs.

Model-RoutingCost-OptimizationAdaptive-RouterComplexity-Gate
GitHub Stars
27 K+
+12.4% this month
Monthly Downloads
105 K+
Monthly registry pulls
Reach Score
98.0/ 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 @yjh051108/dsh-routing-suite

Architecture

DSH Routing Suite resolves the critical dilemma between output quality and infrastructure cost. Sending every trivial greeting or simple regex request to a high-end reasoning model exhausts budgets and increases response latency. This plugin implements an intelligent complexity gatekeeper on Cordis. Incoming user turns undergo micro-second semantic categorization: lightweight classification tasks route to efficient local or small cloud models, while intricate algorithmic planning escalates smoothly to DeepSeek-R1. This hybrid strategy maintains top-tier benchmark quality while trimming cloud spend by up to 60%.

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
Dual-Metric Complexity Scoring: Micro-second evaluation of semantic ambiguity and dependencies
02
Seamless Fallback Escalation: Auto-promotes turns to flagship models if small models fail validation
03
Fine-Grained Budget Caps: Set hard expenditure ceilings per session or billing cycle easily
04
Live Traffic Analytics Dashboard: Real-time visual tracking of token savings and tier distributions

Core Workflow

01

Complexity Scoring

Calculates complexity scores based on prompt tokens, ambiguity, and tool schemas.

02

Rule-Based Matching

Maps evaluated scores to configured model tiers (Tier-1, Tier-2, or Reasoning-Tier).

03

Payload Forwarding

Adjusts temperature and token parameters per target provider before dispatch.

04

Quality Gate Fallback

Transparently escalates to flagship models if small model outputs fail verification.

Configuration Parameters Reference (YAML / JSON)

ParameterTypeDefaultDescription
budgetModestringbalancedRouting strategy prioritization
complexityThresholdnumber0.65Threshold to escalate to DeepSeek-R1
lightweightModelstringdeepseek-chat-fastTarget lightweight model endpoint

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